This article is mentioned in:

Introduction

Brain arteriovenous malformations (bAVMs) were first described ~200 years ago and are defined as the direct communication of arteries to abnormally tortuous and dilated veins without interposing capillaries (1,2). bAVMs may present as hemorrhage, epilepsy, headache, non-hemorrhagic neurological deficits or are asymptomatic. Although it has been associated with certain well-defined genetic disorders, including ataxia telangiectasia, Osler-Weber-Rendu syndrome, Sturge-Weber syndrome and Wyburn-Mason syndrome (3–6), the molecular etiology of bAVM has not been fully elucidated. A previous study has suggested that bAVM may be congenital but dynamic with the possibility to grow, regress and even reappear (7). Several studies have found that there is an association between the upregulation of the Notch signaling pathway and bAVM (8–11). However, whether the pathway is involved in the hemorrhage of bAVMs, has yet to be fully elucidated.

The Notch signaling pathway is a transmembrane, evolutionarily conserved intercellular signaling pathway in vertebrates and non-vertebrates. It affects a wide variety of developmental processes and cell-fate determination during the embryonic period and postnatally (12–14). The Notch signaling pathway consists of four receptors (Notch1–4) and five ligands [Jagged 1 and 2, Delta-like (DLL) 1, 3 and 4] in mammals. It controls lymphatic endothelial quiescence, endothelial cell (EC) selection and the morphogenesis of vascular branches, as well as the interaction between the developing endothelium and endothelial basement membrane and arterial maturation (15–19). It has also been reported to coordinate with vascular endothelial growth factor (VEGF), transforming growth factor-β (TGF-β), the anaplastic lymphoma kinase signaling pathway, the Wnt/β-catenin pathway and nuclear factor-κB (16,17,20–23). The pathway is initiated with the binding of adjacent signaling cell ligands and receptors on receiving cells, followed by two proteolytic cleavage events. The second cleavage event releases the Notch intracellular domain (NICD), which is then translocated to the nucleus and forms a complex with DNA-binding proteins, including CBF1, Su(H) and LAG-1(CSL), displaces previously bound co-repressors and recruits co-activators. In addition, the Notch signaling pathway functions as a transcriptional activator of the downstream targets, hairy and enhancer of split (Hes) and Hes-related protein gene families (24–26). Furthermore, the Notch pathway has been found to be associated with syndromes presenting with abnormal vascular phenotypes, including Alagille (27) and Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy syndromes (28,29).

The present study aimed to assess the expression levels of NOTCH1, NOTCH4, DLL4 and JAGGED1 in the nidus of patients with bAVM by quantitative polymerase chain reaction (qPCR) and immunohistochemical staining to determine whether there is an association between the Notch signaling pathway receptor or ligand expression levels and the clinical hemorrhage of bAVM.

Materials and methods

Subjects

A total of 35 nidus tissues from patients with bAVM were obtained from the Capital Medical University Affiliated Beijing Tiantan Hospital (Beijing, China) from June 2011 to June 2012. All patients were diagnosed with bAVM according to digital subtraction angiography and magnetic resonance imaging results combined with postoperative pathology diagnosis. The patients ranged in age from 7 to 57 years with a mean age of 29.29±14.64 years (median ± standard deviation). The majority of patients presented with the clinical symptoms of bAVM, including intracranial hemorrhage, epilepsy and headaches. Only one patient was recruited via regular medical examination. Among the 35 patients, five underwent endovascular treatment prior to surgery. The patients with bAVM were further divided into two groups: The hemorrhage group and the non-hemorrhage group, based on their first clinical presentation. The mean age of the hemorrhage and non-hemorrhage groups were 25.68±13.38 and 33.56±15.12 years, respectively. The clinical features of patients with bAVM are shown in Table I. Ten normal human superficial temporal arteries (STAs) were obtained from head trauma patients with a mean age of 45.7±16.3 years (range, 23–60 years). Informed consent was obtained from all patients and controls, either directly from the individual or their legal guardian. The protocol of the present study was approved by the ethics committee of the Beijing Tiantan Hospital (Beijing, China).

Table I Clinical features of patients with brain arteriovenous malformation.

Table I

Clinical features of patients with brain arteriovenous malformation.

Clinical variables	Values
Age, years	30.10±14.97 (range, 8–57)
Gender, cases (%)
Male	19 (54.29)
Female	16 (45.71)
Clinical presentation, cases (%)
Epilepsy	17 (48.57)
Hemorrhage	16 (45.71)
Headache	5 (14.29)
Asymptomatic	1 (2.86)
Other cerebarl vascular diseases, cases
Aneurysm	3
Ateriovenous fistula	1
Achnoid cyst	1
Prior embolism, cases (%)	5 (14.29)
S-M score	2.46±0.65

[i] S-M score, Spetzler-Martin score. Values are expressed as the median ± standard deviation.

qPCR

Expression levels of mRNA were normalized against hypoxanthine phosphoribosyl-transferase-1 (HPRT-1), using the following specific PCR primers designed by the NCBI Basic Local Alignment Search Tool (http://blast.ncbi.nlm.nih.gov/Blast.cgi, 2013); primers are shown in Table II. Total RNA was isolated using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA), and cDNA was acquired according to the Takara procedure (Takara Bio, Inc., Shiga, Japan) with 2 μg of total RNA. qPCR reactions for the four genes, NOTCH1, NOTCH4, DLL4 and JAGGED1, were performed using the Takara Thermal Cycler Dice Real-Time System (TP800; Takara Bio, Inc.) and the conditions were as follows: 95°C for 30 sec; 40 cycles comprising 95°C for 5 sec, 55°C for 30 sec, 72°C for 30 sec; and a final hold at 95°C for 15 sec, 60°C for 30 sec and 95°C for 15 sec. All reactions were performed in triplicate. After completion of qPCR, the amplification products underwent electrophoresis immediately on agarose gel. The target gene was confirmed according to the position in the DNA marker. Absorbance data were collected at the end of every extension, and cycle threshold (Ct) values were analyzed by the ΔCt method (30). The efficiency of amplification of the target genes and the internal control (HPRT-1) were examined.

Table II Primers designed for quantitative polymerase chain reaction.

Table II

Primers designed for quantitative polymerase chain reaction.

Genes	Forward primers	Reverse primers
HPRT1	GCCCTGGCGTCGTGATTAGT	GGGCTACAATGTGATGGCCTCC
NOTCH1	GCCGCAGTTGTGCTCCTGAA	TGTCGTGATGCATGCGCTCC
NOTCH4	TGGAGAAGGGGCTGTGGAATG	CACACTGGCAGGTCCCTTGT
DLL4	CGCAATGACCACTTCGGCCA	ATTCGTTACACAGCCGGCCC
JAGGED1	TCGAGTCTGAGGCCGTTGCT	GAGACCGTGTCGGCTGCAAG

[i] HPRT1, hypoxanthine phosphoribosyl-transferase-1; DLL, delta-like.

Immunohistochemistry

Brain AVM specimens and STAs were embedded in paraffin and cut into 5-μm sections. Specimens were deparaffinized with xylene and rehydrated with ethanol following antigen retrieval with antigen unmasking solution according to the manufacturer’s instructions (Vector Laboratories, Burlingame, CA, USA). Endogenous peroxidase activity was blocked by incubation with 3% H2O2 at room temperature for 10 min. Specimens were then washed with phosphate-buffered saline (PBS) and incubated in blocking solution for 3 min at a high pressure (pressure cooker; Biocare Medical, Inc., Concord, CA, USA). The primary antibodies used were as follows: Mouse monoclonal anti-Notch1 intracellular domain antibody (MAB5352, 1:300; Millipore, Billerica, MA, USA), rabbit polyclonal anti-Notch4 intracellular domain antibody (sc-5594, 1:100; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), rabbit polyclonal anti-CD31 (ab28364, 1:200; Abcam, Cambridge, MA, USA), rabbit polyclonal anti-Jagged-1 (ab7771, 1:200; Abcam), rabbit polyclonal anti-Hes-1 (ab71559, 1:100; Abcam) and rabbit polyclonal anti-DSL domain of DLL4 (ab7280, 1:150; Abcam). Primary antibodies were added to the blocking buffer and incubated with the tissue sections at 4°C overnight. Sections were then washed with PBS and incubated with horseradish peroxidase-labeled goat anti-rabbit or anti-mouse antibody for 1 h at room temperature. 3,3′-diaminobenzidine solution (Vector Laboratories) was used to obtain a visible reaction product. The staining process was performed blind from the source. A Leica microscope and Leica digital color camera (DM4000; Leica, Mannheim, Germany) were used to examine and photograph the slides, respectively.

Statistical analyses

Statistical analysis was performed using a two sided t-test and one-way analysis of variance (ANOVA). All data are presented as the mean ± standard deviation. Data were analyzed using SPSS software, version 19.0 (SPSS Inc., Chicago, IL, USA). P≤0.05 was considered to indicate a statistically significant difference.

Results

Expression levels of the four genes in bAVM and STA specimens detected by immunohistochemistry

Elevated expression levels of NOTCH1 and NOTCH4 were detected in the hemorrhage bAVM group compared with that of the STA group. HPRT-1 was selected as the internal control for normalization. The expression levels of NOTCH1 and NOTCH4 were elevated by 2.26-fold (P=0.033) and 2.80-fold (P=0.002), respectively; however, the expression of JAGGED1 and DLL4 were slightly different but without significance (P=0.106 and P=0.492, respectively). In order to further examine the ligands involved in the Notch signaling pathway, immunohistochemical analysis was perfomed using a positive control, the middle cerebral artery (MCA). To determine the phenotype of expressing cells, anti-CD31 was selected as a representative. Immunohistological analysis demonstrated that Notch1 and Notch4 were weakly expressed in ECs in MCA and STA (Fig. 2), but overexpressed in the hemorrhage and non-hemorrhage groups. Jagged1 and DLL4 demonstrated similar expression levels in the two control and bAVM groups. The downstream target gene of the Notch signaling pathway, HES1, was predominantly expressed in nuclei of ECs and smooth muscle cells in bAVM, but was negatively expressed in the STAs and MCAs.

Figure 2 IH staining of STA, MCA, hemorrhaged AVM and non-hemorrhaged AVM samples. (A–D) Notch1, (E–H) in STA, MCA, hemorrhaged AVM and non-hemorrhaged AVM, respectively. (E–H) Notch4 in STA, MCA hemorrhaged AVM and non-hemorrhaged AVM. (I–L) Jagged1 in STA, MCA hemorrhaged AVM and non-hemorrhaged AVM. (M–P) IH stain for DLL4. (Q–T) IH stain for Hes1. There were no significant differences between the STA and MCA groups in (A,B) Notch1, (E,F) Notch4, (I,J) Jagged1, (M,N) DLL4, (Q,R) Hes1 and (U,V) CD31. In concordance with the quantitative polymerase chain reaction results, Notch1 and Notch4 expression were significantly increased in the AVM groups, whereas DLL4 and Jagged1 expression were similar in the STA control group. Compared with the hemorrhage bAVM and non-hemorrhage bAVM groups, only Notch1 was upregulated in hemorrhage brain AVM, other components remained unchanged in bAVM groups. (U–X) CD31 was stained to demonstrate endothelial cells. IH, immunohistochemical; STA, superficial temporal artery; MCA, middle cerebral artery; bAVM, brain arteriovenous malformation; DLL, delta-like; Hes-1, hairy and enhancer of split-1.

Expression levels of the four genes detected by qPCR

qPCR detected a significant increase (2.82-fold; 0.629±0.596) in NOTCH1 expression in the hemorrhage group compared with that of the non-hemorrhage group (P=0.023). The expression levels of NOTCH4 and DLL4 remained unchanged between the two groups, while JAGGED1 was overexpressed in the non-hemorrhage group (1.24-fold) (Fig. 1). However, the overexpression of JAGGED1 lacked significance when analyzed by ANOVA. The expression levels detected by qPCR of the key components of the Notch signaling pathway in the bAVM groups were in concordance with the immunohistochemical results. The lack of significant differences in the expression of JAGGED1 between the hemorrhage and non-hemorrhage groups indicated that the abnormal expression levels may be caused by the overall expression levels of JAGGED1 in neural cells.

Figure 1 Relative mRNA expression levels of key factors of the Notch signaling pathway in hemorrhage AVM, non-hemorrhage AVM and STA. (A) NOTCH1 expression levels increased by 2.82-fold in hemorrhaged AVM. (B) NOTCH4 and (C) DLL4 transcription levels remained unchanged. (D) JAGGED1 was overexpressed in the non-hemorrhage group (1.24-fold), but without significance. *P<0.05; **P>0.05. AVM, arteriovenous malformation; STA, superficial temporal artery; DLL, delta-like.

Discussion

bAVMs are commonly detected in patients aged ~30 years and there have been a number of cases reporting the regrowth of bAVM following complete surgery or embolism, reflecting the enigmatic etiology of bAVM (2,31). Due to the scarce data available on these cases, it is not possible to arbitrarily conclude that bAVM is a complete congenital lesion.

Previous studies have suggested that NOTCH1 is an oncogene as its upregulation is associated with the presence of solid tumors in the digestive system, breast and hematological system (32–34). Uyttendaele et al (10) found that the introduction of the active form of NOTCH4 and knockout of either NOTCH1 or NOTCH1/NOTCH4 in ECs produced similar phenotypes of vascular deficiency with the loss of fine branches or excessive branches. In addition, Murphy et al (9) observed the increased activation of endothelial Notch signaling with the presence of bAVM-like structures in mice, which suggested that the activation of the Notch signaling pathway is potentially a crucial molecular candidate in bAVM pathogenesis. In addition, Zhuge et al (11) identified the upregulation of Notch1 in human samples, confirming that abnormal Notch signaling activation has a vital role in the pathogenesis of bAVM. In a further study to elucidate the mechanisms of the Notch signaling pathway and determine potential treatment for bAVM, Murphy et al (9) focused on the normalization of the receptor Notch4. In his study, the high-flow AV shunts, caused by tetracycline-responsive system, decreased in blood flow as well as vessel diameter following repression of Notch4. Based on these findings, therapy that targets the Notch signaling pathway may be a promising approach for bAVM treatment (8,11,35). In addition, a previous study using mice demonstrated that a complete and safe normalization of Notch4 in animals reduced blood vessel size in bAVM, suggesting a treatment option of bAVM alternative to surgery, embolism or radiation therapy (8,35).

In the present study, elevated expression levels of NOTCH1 and NOTCH4 were found in the bAVM nidus at the RNA and protein levels compared with those of normal STA and normal MCA, which was in accordance with previous research (10,11,35). To the best of our knowledge, the present study was the first to determine whether the Notch signaling pathway is associated with the clinical presentation of hemorrhage in bAVM. Patients with bAVM were stratified into the hemorrhage and non-hemorrhage groups and the expression levels of key factors in the Notch signaling pathway were compared between the groups. It was found that the expression levels of NOTCH1 significantly increased in the hemorrhage group, suggesting that NOTCH4, DLL4 and JAGGGED1 may not be involved in the pathogenesis of hemorrhage in bAVM nidus. Recently, Liebler et al (36) reported that while Notch1 NICD markedly repressed endothelial migration and sprouting angiogenesis, the intracellular domain of Jagged1 and DLL4 did not influence EC adhesion. It is therefore concluded that the relevance of ligand intracellular domains in ECs remains unclear. However, our findings also showed different results compared with those of Zhuge et al (11). It was demonstrated that the expression levels of DLL4 and JAGGED1 appeared unchanged between patients with bAVM and the controls. As a large sample size was used to contribute to the power of detection, the data are considered reliable.

In the present study, it was demonstrated that the expression levels of NOTCH1 and NOTCH4 were increased in patients with bAVM, and the correlation between hemorrhage and Notch1/4 expression was analyzed. It was demonstrated that the increased expression of key components of the Notch signaling pathway are a potential cause of hemorrhage in human bAVM. The findings also suggested that, although the continuous activation of the Notch signaling pathway was reported to lead to hemorrhage in the brain, liver or other organs in studies of mice and embryos (8,9), there was no significant association between hemorrhage of bAVM and Notch4, DLL4 and Jagged1; however, there was an association between hemorrhage and Notch1. Further studies targeting the remaining receptors and ligands of the Notch signaling pathway are required to conclude whether Notch1 is the key factor involved in hemorrhage in bAVM. This would elucidate the importance of anti-Notch signaling drugs and their efficiency to reduce the risk of hemorrhage in patients with bAVM.

Abbreviations:

References