Differential expression levels of collagen 1A2, tissue inhibitor of metalloproteinase 4, and cathepsin B in intracranial aneurysms
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.185350
Source of Support: None, Conflict of Interest: None
Aims: Intracranial aneurysms (IAs) express a variety of differentially expressed genes when compared to the normal artery. The aim of this study was to evaluate the expression level of a few genes in the aneurysm wall and to correlate them with various clinicoradiological factors.
Keywords: Cathepsin B; collagen 1A2; intracranial aneurysms; tissue inhibitor of metalloproteinase 4
Subarachnoid hemorrhage (SAH) resulting from the rupture of intracranial aneurysms (IAs) is an important cause of hemorrhagic stroke. The prevalence of IA ranges from 1% to 6% of the world's population. Hemodynamic stress at arterial bifurcations is thought to be the main reason for the formation of aneurysms. Structural weaknesses seen in connective tissue disorders such as autosomal dominant polycystic kidney disease and Ehlers–Danlos syndrome are associated with the presence of IAs. Familial occurrence of IAs suggests that there are genetic factors involved in the development of aneurysms with the risk being 3–7 times higher in first-degree relatives of patients with SAH than in the general population.,,,, Whole genome screening with linkage analysis has been used in the past to find the involved gene locus in patients with aneurysms.
In the present study, the first of its kind, we have evaluated the gene expression levels of cathepsin B (CTSB), collagen 1A2 (COL1A2), and tissue inhibitor of metalloproteinase 4 (TIMP4) in the intracranial aneurysmal walls compared to the normal superficial temporal artery (STA) taken from the same patient. The gene expression assessment was performed by the real-time polymerase chain reaction (RT-PCR). This is the first study that has been performed on Indian patients and also the first to try and find an association of their gene expression levels with various other clinical and radiological factors. The latter parameters have not been analyzed in the studies in the available literature.
Patients and tissue samples
Over the study period from September 2011 to November 2013, 23 samples of aneurysmal sac and 19 samples of the STA were taken from 23 patients undergoing craniotomy for clipping of IAs after obtaining consent for the study. Adult patients of the age range 18–75 years, diagnosed with sporadic IA were included in the study. Patients with a family history of IA and/or SAH in their first-degree relatives, patients with known genetic disorders such as polycystic kidney disease and Marfan's syndrome, and patients with complex IAs in whom the excision of the sac was considered difficult were excluded from the study. The Institutional Scientific Ethics Committee approved the study and ethical clearance and patient consent were obtained before starting the sample collection.
Collection and storage of tissue samples
During the surgery, after clipping of the aneurysm, if there was a favorable anatomy, the wall of the aneurysmal sac (fundus of the aneurysm) was excised. A branch of the STA was also harvested from the same patient at the time of closure of the wound, whenever possible. The samples were transferred into small containers containing RNAlater solution (ThermoFisher Scientific, MA, USA), the containers were immediately placed at 4°C and RNA isolation was carried out. When a longer storage was required, samples were transferred to −80°C within 24 h. An additional replicative set of 17 ruptured aneurysm samples and 7 STA samples was included for validation.
Isolation of RNA from tissue samples
RNA was isolated from tissue samples with Applied Biosystem's RNA PureLink Mini Kit. All the reactions were carried out in an ESCO Biological Safety Cabinet with laminar airflow. The quantity and quality of RNA were analyzed with NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DW, USA). Only good-quality RNA with an A260/A280 ratio of more than 1.8 was used.
Real-time polymerase chain reaction
RT-PCR was done using the two-step method. Five hundred micrograms of total RNA from each sample was converted to cDNA with Applied Biosystems High Capacity cDNA Reverse Transcription Kit in a 20 µl reaction. TaqMan ® Gene Expression Assays were used for RT-PCR. HPRT1 (Hs02800695_m1, VIC) gene, a housekeeping gene, involved in DNA synthesis, was used as an endogenous control to correct for differences in the RNA quantity. Duplex PCR was used and typically one probe was used to detect the target gene, i.e., COL1A2 (Hs00164099_m1. FAM), TIMP4 (Hs00162784_m1, FAM), CTSB (Hs00947439_m1, FAM); another probe was used to detect an endogenous control (reference gene), i.e., the HPRT1 gene.
Analysis of the result
The efficiency of the TaqMan gene expression assays was calculated by plotting the threshold cycle (CT) value against the serial dilutions (at least 4) of a particular gene in duplex reaction with the housekeeping gene (HGPRT1).
Ratio (fold change) = (Etarget gene)ΔCT (target)/(Ereference gene)ΔCT (reference)
i.e., Ratio (fold change) = (Eaneurysm)ΔCT (target-HGPRT)/(ESTA)ΔCT (target-HGPR)
The fold change of each gene studied (CTSB, TIMP4, COL 1A2) was computed individually for each aneurysm sample taken from the patients by comparing CT values of aneurysmal sac and the STA biopsied from the same patient. For the patients from whom only the aneurysmal sac could be harvested, the mean computed CT value of the STA samples was used to compute the fold change of the genes studied.
All data were entered into a database and analyzed using SPSS 15.0 software, (IBM, USA) package. All the variables were tested for normality with Shapiro–Wilk tests. None of the variables were found to be normally distributed. Categorical variables were analyzed with the chi-square test or the Fisher exact test when appropriate. Continuous variables and categorical variables were compared using the Mann–Whitney U-test or Kruskal–Wallis test as appropriate. To identify the association between the gene expression and various other continuous variables, Spearman's correlation was used. The level of significance was fixed at 5%. All P values are two-sided and P ≤ 0.05 was considered statistically significant.
Demographics and clinical features
Twenty-three patients were included in the study over the period from September 2011 to November 2013. The mean age of the patients was 51.04 ± 13.85 years (median – 55 years; range: 24–75 years). There were 10 male and 13 female patients included in the study, with a male–female ratio of 1:1.3.
Twenty-two out of 23 patients (95.6%) presented with SAH due to rupture of the aneurysm. One patient presented with features of raised intracranial pressure (ICP) with generalized tonic-clonic seizures due to an unruptured giant thrombosed right middle cerebral artery (MCA) aneurysm. One patient presented with recurrent transient right hemiparesis due to a large left MCA aneurysm. A family history of stroke was present in one patient. The details of the clinical features are given in [Table 1].
All the patients underwent plain CT scan of the head at the time of admission. Of the 23 patients, 18 (78.3%) had Fisher Grade III SAH and 4/23 (17.4%) patients had Grade IV SAH [Table 1]. The details of radiological features are given in [Table 2]. In the patient who had multiple aneurysms, the left MCA bifurcation was the ruptured aneurysm, and he underwent clipping of the same.
Management and outcome
Three out of 23 (13%) patients underwent external ventricular drainage for acute hydrocephalus. All the patients underwent either a standard pterional craniotomy or a larger frontotemporal craniotomy for clipping of the aneurysm.
There were two patients with major intraoperative complications. In one patient with a right ophthalmic segment internal cerebral artery aneurysm, the neck of the aneurysm was avulsed from the parent artery while applying the permanent clip. In another patient with a right MCA aneurysm, the parent artery was occluded with the permanent clip. Fourteen out of 23 (60.8%) patients developed new postoperative deficits or worsening of the previously existing deficits. Of these, 7 (30.4%) patients had transient neurological deficits, which improved with hypertension, hypervolemia, and hemodilution therapy, and the remaining seven patients ended up having permanent deficits. Vasospasm was documented in the postoperative angiograms in 7/12 patients. There were 2 (8.7%) in-hospital mortalities. The cause of death in one of the patients was severe diffuse vasospasm and the other patient died due to myocardial infarction. Follow-up was available for 11 patients (47.8%) (median – 1 month; range: 1–18 months). According to the Glasgow outcome scale (GOS), 9/11 patients had a good recovery; one patient had moderate disability; and the other patient had severe disability at follow-up [Table 3].
The average quantity of RNA extracted from the aneurysm samples varied with the quantity of available sample/size of the aneurysm and was 28.85 ± 21.42 ng/ul (9–92 ng/ul) and the mean RNA quality of aneurysm (260/280 nm value) was 1.99 ± 0.11 (range: 1.83–2.38). The average RNA quantity of STA sample was 44.35 ± 25.53 ng/ul (range: 16–115 ng/ul). The mean RNA quality of STA samples (260/280 nm value) was 1.97 ± 0.11 (range: 1.58–2.08).
Fold change of the genes
Data from aneurysm samples of two patients could not be ascertained due to technical reasons; however, the values for the corresponding STA samples were obtained and were included in the computation for the average CT value of STA samples and utilized in the analysis.
The mean fold change of COL1A2 between the aneurysm sample and the STA sample was 2.46 ± 0.12 (range: 0.12–8.91); i.e., COL1A2 gene was 2.46 times overexpressed in the aneurysmal sac compared to the STA on average.
The mean fold change of TIMP4 between the aneurysm sample and the STA sample was 0.31 ± 0.21 (0.10–0.84), i.e., TIMP4 gene was 3.23 times underexpressed in the aneurysmal sac compared to the STA on average.
The mean fold change of CTSB between the aneurysm sample and the STA sample was 31.47 ± 39.01 (range: 4.99–138.09), i.e., CTSB gene was 31.47 times overexpressed in the aneurysmal sac compared to the STA on average.
An additional 17 ruptured aneurysm samples were analyzed for validation, which also showed similar trends of differential expression, with COL1A2 and CTSB being overexpressed and TIMP4 underexpressed in the aneurysmal sac compared to the STA.
Association of the gene expression and various other factors
Fold changes of TIMP4, COL1A2, and CTSB were compared with various clinical and radiological variables. There was a good correlation or association between the following variables:
In the present study, we have studied the mRNA expression of COL1A2, TIMP4, and CTSB genes in the aneurysmal sac and compared that with the average expression level of these genes in the normal superficial artery taken from the same patient group. When we compared the fold change of these genes between the aneurysmal sac and the normal artery with various clinical and radiological factors, we found some interesting results.
In the previous studies of gene expression profiling of IA using DNA microarray, COL1A2, TIMP4, and CTSB were consistently differentially expressed in the aneurysmal wall compared to the normal artery. In this study, we have replicated the results of previous studies using RT-PCR and correlated the expression of these genes with various clinical and radiological factors. There are a few studies of whole genome expression profiling using DNA microarray technology available in the literature [Table 4]. In a meta-analysis of five such studies on IAs by Roder et al.,,,,,, a total of 507 genes with altered expression were listed, of which 57 showed differences in more than two studies and seven in more than three studies (BCL2, COL1A2, COL3A1, COL5A2, CXCL12, TIMP4, and TNC). Aoki et al., demonstrated upregulated expression of CTSB, K, and S in the late stage of aneurysm progression in rats and humans, and treatment with NC-2300 (cathepsin inhibitor) resulted in the decreased incidence of advanced aneurysms. CTSB was also upregulated in two of the five studies taken up for meta-analysis by Roder et al.
Upregulation of different types of collagen have been described in many previous gene expression studies. Shi et al., found that the COL1A2 gene is 4.72 times overexpressed in the aneurysm wall compared to the STA. Other gene expression studies by Li et al., Marchese et al., and Krischek et al. have also shown similar results.,, In the present study, COL1A2 expression level was studied using RT-PCR and the gene was 2.46 times overexpressed in the aneurysm wall compared to the superficial artery taken from the same patient.
In a weak arterial wall as seen in IAs, expression of collagen is expected to be low. The higher expression that has been consistently reported might reflect increased collagen mRNA synthesis to compensate for a higher collagen degradation caused by increased gelatinase activity, inflammatory cells, or higher activity of matrix metalloproteinases (MMPs), as shown in other studies.,
Matrix metalloproteinases and tissue inhibitor of metalloproteinases in the development of intracranial aneurysms
Matrix metalloproteinases (MMPs) and their endogenous inhibitors, the TIMPs, modulate the extracellular matrix (ECM) components., Pera et al., in their gene expression study, have demonstrated that fold change of TIMP4, compared to middle meningeal artery, was 0.45 in ruptured aneurysms and 0.43 in unruptured aneurysms. Marchese et al., and Krischek et al., have also shown similar results.,, In the present study, the fold change of TIMP4 was 0.31 (3.22 times underexpressed) in the aneurysm wall compared to the normal STA. Decreased expression of TIMP4 would have resulted in increased activity of MMPs, especially MMP2 and MMP9, which are known to have increased expression in IAs, resulting in uncontrolled ECM remodeling and subsequent development of IA.
In the present study, there was a significant positive correlation of TIMP4 expression with the maximum diameter and the size of the fundus of the aneurysm. None of the previous studies have demonstrated this kind of correlation. Increased expression of TIMP4 in the aneurysm wall would be expected to lead to decreased amounts of MMPs, and hence, a stronger arterial wall. The larger size of the aneurysm with an increased TIMP4 expression might indicate compensational mechanisms due to higher TIMP degradation.
Previous studies have shown that there is increased expression of MMPs and decreased expression of their inhibitors in gestational and systemic hypertension., In the present study, there was significant negative correlation of TIMP4 expression level and the systolic BP at admission, though there was no significant association between the presence of hypertension and the TIMP4 expression level. The increased systolic BP at admission could be either due to undetected previous systemic hypertension or a compensatory increase in the BP due to raised intracranial pressure after SAH.
Cathepsins consist of large family members of lysosomal proteolytic enzymes and CTSB is a cysteine cathepsin. Shi et al., found that the CTSB gene is 6.50 times overexpressed in the aneurysm wall compared to the STA. Marchese et al., and Aoki et al., also showed that the CTSB is overexpressed in the aneurysmal wall compared to the normal artery., In the present study, CTSB was 31.47 times overexpressed in the aneurysmal sac compared to STA, which is the highest change observed among all the three genes studied in the present study. Overexpression of the CTSB could have led to increased protein degradation of the ECM, leading to a weaker arterial wall and thus resulting in the formation and progression of the aneurysm.
The expression of CTSB in the aneurysmal wall was significantly low in the patients who developed fresh postoperative deficits compared to those who did not (P = 0.039). This can be explained by an assumption that there would have been a relatively decreased expression of the CTSB in all the intracranial arteries in patients who had fresh postoperative deficits (although this was not evaluated in the present study). Since CTSB is a lysosomal enzyme, its decreased expression in the intracranial arteries would have led to decreased degradation of some inflammatory and pro-vasospasm cytokines, which would, in turn, have led into increased incidence of vasospasm, and hence to fresh postoperative deficits.
The patients who had hydrocephalus at presentation had higher levels of CTSB expression in the aneurysmal wall compared to those who did not have hydrocephalus (P = 0.008). This can be explained by the higher level of expression of CTSB in patients who presented with Fisher grade 3 SAH (37.18) compared to those who presented with Fisher grade 4 SAH (10.00) [though this was not statistically significant with P = 0.057]. Thus, the increased blood in the subarachnoid space in the Fisher grade III SAH led to a decreased CSF absorption, resulting in increased incidence of hydrocephalus at presentation.
In the present study, we have shown differential expression of COL1A2, TIMP4, and CTSB in the aneurysmal wall compared to the normal artery, similar to the previous studies published so far [Table 4]. We have also shown the association or correlation of various clinicoradiological factors with the expression level, which has not been reported in the previous studies.
Since only three of the many genes were evaluated in the present study, it is difficult to draw conclusions regarding the genetic origin of the IAs. However, the study is the first of its kind done in the Indian population and has definitely provided an insight into the pathogenesis of the aneurysms and has unveiled the association of a few clinicoradiological factors with gene expression, which has not been studied previously. In the present study, the mRNA expression was studied with RT-PCR, which is a more focused approach of looking into gene expression than the whole genome analysis using the DNA microarray technology.
Limitations of the study and future directions
The present study is one of the largest gene expression studies of the intracranial aneurysmal wall and the first of its kind done in the Indian population. Yet, the findings need to be interpreted with caution because of the limitations.
From over 400 patients operated over the study period, only forty patients were selected according to the aneurysm morphology (rather than selecting the subjects randomly), and this has the limitation of a selection bias. Second, gene expression was compared between the aneurysmal wall and the normal artery from the same patient. Studying the gene expression between the aneurysm wall and the normal artery from a different set of population will give further insights into the pathogenesis of the disease. Finally, we have not studied utilizing immunohistochemistry or proteomics whether differential expressions of these genes were translated to the target proteins.
Whether this differential expression of the studied genes led to aneurysm formation or if this was only due to hemodynamic changes in the arterial wall following the aneurysm formation needs to be elucidated in future studies. In the future, the results of genome-wide association studies (GWAS) target sequencing studies of the locus identified by the GWAS, gene expression studies, and exome sequencing studies can be analyzed together and the genes involved in the formation and progression of the aneurysm can be shortlisted and then targeted for gene therapy.
COL1A2 gene and cathepsin genes were overexpressed and TIMP4 gene was underexpressed in the aneurysmal sac compared to the STA, and interestingly, we found the expression levels were associated with a few clinicoradiological factors. Since the study sample size was small, we cannot conclude anything substantial regarding the correlation of the genetic expression with the clinicoradiological factors. Further studies are required to elucidate whether or not this differential expression of RNA is also reflected in the protein expression of the studied genes; and, if this differential expression of the studied genes has led to aneurysm formation or if this differential expression of genes was only due to the hemodynamic changes in the arterial wall following the aneurysm formation.
We thank Mr. Prem K. and Mr. Chandrashekar MR, lab technicians, Department of Neuropathology, for their contributions in collecting and processing the samples in the laboratory.
Awards: This paper was selected as among the top International Abstracts and awarded the AANS International Travel Scholarship at American Association of Neurological Surgeons Annual Conference – 2015.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4]