Although intracranial aneurysms (IAs) are a major public health problem in the United States, few etiological factors are known. Most aneurysms remain asymptomatic until they rupture, producing subarachnoid hemorrhage, one of the most severe forms of stroke. Despite the technical advances in endovascular and microsurgical treatment, these patients still have high mortality and morbidity rates. Hence, the biology of aneurysm formation and growth is of intense interest. The presence of T and B lymphocytes, as well as macrophages, in human IA tissues suggests a role for inflammation in IA pathogenesis. However, the types of cytokines that are involved and regulated during cerebral aneurysm formation and growth are not known. To study the underlying pathogenesis of IA, we analyzed the expression of cytokines that participate in proinflammatory and anti-inflammatory responses.
Polymerase chain reaction was used to assess relative messenger ribonucleic acid expression levels of cytokines and an apoptotic modulator, Fas-associated death domain protein. Western blot analysis was used to determine protein expression from these genes.
We show that the proinflammatory cytokine, tumor necrosis factor α and its proapoptotic downstream target, Fas-associated death domain protein, are increased in human aneurysms. In contrast, interleukin 10, which is secreted predominantly by T helper 2 cells, was absent in aneurysms. Polymerase chain reaction-derived gene expression data were confirmed by Western blotting using specific antibodies.
Increased tumor necrosis factor α and Fas-associated death domain protein may have deleterious primary and secondary effects on cerebral arteries by promoting inflammation and subsequent apoptosis in vascular and immune cells, thereby weakening vessel walls.
It is estimated that between 2 and 10 million adults in the United States harbor cerebral aneurysms (5). This deadly disease occurs when a bulge in a blood vessel wall forms where the arteries branch at the base of the brain. Although the cause of aneurysms remains a mystery, hypertension and smoking have been suggested as risk factors that contribute to the onset of the disease (3,13). Aneurysms begin in a harmless form without symptoms. However, some undergo progressive inflammation and eventually rupture, causing a hemorrhage. Each year, more than 30,000 persons with cerebral aneurysms experience a hemorrhage, with a survival rate of only 25%. Strokes are common and are one of the consequences of the excess bleeding (6,14). The best way to treat these patients is to detect aneurysms before they rupture. Several new technologies have been developed to stop aneurysm growth and rupture, including nonsurgical procedures that use angiograms to insert coils into blood vessels to block the aneurysm. Although these procedures work well to prevent rupture, a fundamental understanding of the disease formation, growth, and rupture could lead to the development of therapeutics to prevent these processes.
In healthy humans, inflammation occurs as a natural response to injury. Inflammation includes swelling, redness, and increased infiltration of white blood cells. On injury, inflammation is beneficial, facilitating the initial repair of damaged tissue. When persistent, however, inflammation may have deleterious effects (12). Histopathological studies on clipped human aneurysms have identified macrophages and lymphocytes in the aneurysm wall; these cell types are indicators of inflammation (9,10). Although the balance between proinflammatory and anti-inflammatory cytokines is likely to determine the level and persistence of inflammation, the contribution of these factors to aneurysm-related inflammation is not known. Tumor necrosis factor α (TNFα) and interleukin-10 (IL-10) are known to exert proinflammatory and anti-inflammatory activities, respectively. TNFα not only elicits inflammation but also serves as an inductive signal to initiate apoptosis. An analysis of cells derived from Fas-associated death domain (FADD) protein knockout animals (21) suggests a critical role for FADD protein in TNFα-induced apoptosis. Therefore, we examined inflammatory and proapoptotic mediators that may contribute to the formation and rupture of human aneurysms.
PATIENTS AND METHODS
Between January 1999 and August 2004, 353 patients with intracranial aneurysms (IAs) (304 with associated subarachnoid hemorrhage and 49 without subarachnoid hemorrhage) were treated microsurgically in the Neurosurgery Department of Marmara University in Istanbul, Turkey. A total of 422 aneurysms were clipped. In 10 of these 353 patients, we collected aneurysm wall specimens from ruptured aneurysms (Table 1). Of these patients, five were men with a median age of 49.6 years (range, 36–62 yr) and five were women with a median age of 53.8 years (range, 47–63 yr). We sampled the outer one-third to one-half of the aneurysms, including the domes. All of the operations were performed within the first 5 days after subarachnoid hemorrhage was diagnosed, and all of the specimens were from berry aneurysms smaller than 2.5 cm. The tissue samples were stored in liquid nitrogen for molecular biology studies. The distal portion of the superficial temporal artery was used as a control artery. All of the 10 individuals included in the study were Turkish.
Ribonucleic Acid Extraction
All the tissue specimens were transported from Turkey to the United States on dry ice and, on arrival, were stored immediately at −80°C. Total ribonucleic acid was extracted using Trizol (Invitrogen Corp., Carlsbad, CA), according to the manufacturer's instructions. The same ribonucleic acid extraction protocol was used for control and aneurysm samples.
Semiquantitative ReverseTranscriptase-Polymerase Chain Reaction
Total ribonucleic acid (1 μg) from each sample was reverse transcribed with oligo(dT) using the SuperScript First-Strand Synthesis System (Invitrogen). Amplifications of 32 to 35 cycles were performed in an automated thermocycler (denaturation, 94°C for 60 s; annealing, 58°C for 60 s; extension, 72°C for 60 s). The amplified polymerase chain reaction products were electrophoresed on a 2.0% agarose gel and were visualized with ethidium bromide/ultraviolet light. The primers used in this study were as follows: an internal control, glyceraldehyde 3-phosphate dehydrogenase (500 base pairs [bp]): forward, GAAGGTGAAGGTCGGAGTC; reverse, CAAAGTTGTCATGGATGACC; TNFα (311 bp): forward, AGGCGCCACCACGCTCTTCT; reverse, GGCAGCCTTGGCCCTTGAA; IL-10 (223 bp): forward, CAACCTGCCTAACATGCTTCGAG; reverse, GAGTTCACAGCGCCTTGATGTC; FADD protein (205 bp): forward, GGAAGAAGACCTGTGTGCA; reverse, CAGGTGGGCCACTGTTGC. Quantitation of polymerase chain reaction products was performed using National Institutes of Health image software (National Institutes of Health, Bethesda, MD). Results were normalized to human glyceraldehyde 3-phosphate dehydrogenase amplified from the same complementary deoxyribonucleic acid mix.
Western Blot Analysis
Control and aneurysm tissues were homogenized in ice-cold lysis buffer containing 0.5% Nonidet P-40 (vol/vol), 25 mmol/L N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid, pH 7.4, 150 mmol/L NaCl, 1 mmol/L ethylenediamine tetra-acetic acid, 1 μmol/L sodium orthovanadate, and a cocktail of protease inhibitors (Calbiochem, San Diego, CA). Homogenates were centrifuged at 13,000 × g in a microcentrifuge at 4°C, and the protein concentrations were determined from the supernatants using the Bio-Rad protein assay (Bio-Rad, Hercules, CA). Proteins (200 μg) from control and aneurysm samples were subjected to 15% sodium dodecyl (lauryl) sulfate-polyacrylamide gel electrophoresis and immunoblotting. The membranes were blocked for 1 hour in Tris-buffered saline with Tween 20 (20 mmol/L Tris HCl, pH 7.4, 0.9% NaCl, and 0.05% Tween 20) containing 5% nonfat dried milk, followed by incubation with primary antibodies against TNFα (Oncogene Research Products, Cambridge, MA), IL-10 (Sigma, St. Louis, MO), or FADD protein (Calbiochem). After extensive washing, the membranes were incubated for 1 hour at 4°C with the appropriate secondary antibody (goat antirabbit immunoglobulin G [Pharmingen, San Diego, CA] or goat antimouse immunoglobulin G [Santa Cruz Biotechnology, Santa Cruz, CA]) conjugated to horseradish peroxidase in Tris-buffered saline with Tween 20 containing 5% nonfat dried milk. The immunoblots were analyzed using an enhanced chemiluminescence detection system (Amersham, Piscataway, NJ).
Values are expressed as means ± standard deviation. Proinflammatory and proapoptotic marker genes and proteins were compared from three independent experiments and were analyzed statistically using the Student's t test for paired data. Differences were considered statistically significant at P < 0.05.
Higher Expression of TNFα in Human Aneurysms
One of the hallmarks of inflammation is the presence of proinflammatory cell types and their mediators. Because TNFα is expressed in various inflammatory conditions, we examined the expression of its messenger ribonucleic acid (mRNA) in clipped human aneurysms and in the superficial temporal artery (control cerebral artery). TNFα mRNA was increased four- to sixfold in aneurysm samples as compared with control arteries (Fig. 1). These results suggest that the proinflammatory cytokine TNFα is significantly increased in human cerebral aneurysms.
Lack of IL-10 Expression in Human Aneurysms
IL-10 is a pluripotent, anti-inflammatory cytokine that affects numerous cell populations, particularly circulating and resident immune cells. One of the biological functions of IL-10 is the limitation and termination of inflammatory responses. Because we observed higher TNFα expression in cerebral aneurysms, we determined whether the expression of IL-10 mRNA is either reduced or absent in human aneurysms. Although IL-10 mRNA was detected in the positive control (Fig. 2C), none of the aneurysm samples expressed this transcript (Fig. 2, A–C). These results suggest that IL-10 expression is either absent or suppressed in human aneurysms.
Increased FADD Protein Expression in Human Aneurysms
The FADD protein, which is a component of the death-inducing signaling complex, is recruited to death-inducing signaling complex in response to death receptor-mediated signaling. FADD protein is a common conduit in both Fas-mediated and TNF receptor (TNF-R)-mediated apoptosis. As illustrated in Figure 3, the expression of the FADD protein-specific transcript was 5- to 40-fold higher in aneurysm samples compared with control superficial temporal arteries. This up-regulation of FADD protein expression may constitute a basis for TNF-mediated apoptosis, as observed in human IA (16).
The Levels of TNFα and FADD protein, but not IL-10, are Elevated in Human Aneurysms
To confirm that increased TNFα and FADD protein mRNA levels in aneurysm tissues correlate with increased protein expression, we performed Western blot analysis with specific antibodies that recognize TNFα and FADD protein. TNFα and FADD protein levels were significantly higher in aneurysms as compared with the control (Fig. 4A). In contrast, IL-10 could not be detected using an IL-10-specific antibody. These results confirm that TNFα and FADD proteins also are significantly increased in aneurysms (Fig. 4B).
Our study has generated three important findings. First, TNFα is increased significantly in human aneurysms as compared with control arteries, extending the functional implications of earlier findings showing that activated monocytes and macrophages and T lymphocytes are present in aneurysms (9,10). Second, IL-10 expression is absent in aneurysms, suggesting the predominance of T helper 1 lymphocyte activation. Finally, FADD protein, an important downstream component of the TNF/TNF-R apoptotic pathway, is increased in aneurysms, indicating that apoptosis may contribute to aneurysm formation and growth.
It is well documented in inflammatory disorders that proinflammatory and anti-inflammatory cytokines facilitate or inhibit, respectively, the activation of various types of cells, including monocytes and macrophages, neutrophils, and T cells, all of which are involved in local inflammation (2,4). In this context, the higher expression of TNFα in human IA lends support to these findings and points to activation of its downstream signaling components to initiate vessel weakening and remodeling via inflammation and apoptosis, as illustrated in Figure 5. Accordingly, TNFα can stimulate cells that produce other vascular inflammatory mediators, such as cytokines IL-1 and IL-6, metalloproteinases, and adhesion molecules. Although we have found no correlation between age, sex, and racial differences to TNFα expression in this study, differential expression of TNFα in ruptured versus nonruptured aneurysms remains a possibility. Additional future experiments will determine this possibility. It is tempting to hypothesize that abnormal and prolonged TNFα expression not only may sustain inflammation, but also increase the activation of degradative enzymes, leading to vessel wall rupture in aneurysms. If this is the case, then limiting the level of TNFα may be beneficial in preventing aneurysm growth and rupture.
TNFα activation also may have other consequences, such as the induction of caspases, which are critical for apoptotic processes (1,8,20). Studies with gene-targeted knockout mice lacking either of two TNF-R forms, TNF-Rp55 or TNF-Rp75, have shown that TNF-Rp55 is important for TNF-induced apoptosis (19). TNF-Rp55 shares homology with Fas, another member of the TNF-R family, and contains an intracellular death domain, FADD protein. The FADD protein acts as an adapter in both TNF-Rp55-associated and Fas-associated death-inducing cascades and is responsible for downstream signal transduction by recruiting and activating caspases. Our data show that the FADD protein is highly expressed in human aneurysms, and thus it may be involved in apoptotic signaling. Although apoptosis has been suggested to occur in human IA walls (11,16,18), to our knowledge, our data are the first to suggest the involvement of the TNFα-regulated apoptotic pathway in aneurysm progression.
The resolution of inflammation is a highly coordinated and active process that is controlled by endogenous proresolving mediators (7,15) such as IL-10, the principal anti-inflammatory cytokine (4). IL-10 is produced by activated T and B cells, monocytes, macrophages, and other immune cells. Its key biological activities include suppression of both lymphocyte and macrophage activation and proinflammatory cytokine production (2). There have been numerous experimental studies of IL-10 in various pathological models, but the number of clinical reports focusing on the profile of this cytokine in humans is limited, especially with regard to neurovascular abnormalities. The lack of IL-10 expression in human aneurysms suggests that T helper 2 cells are either absent or inactivated. Further studies are required to resolve these possibilities.
The relatively small sample size (n = 10) and the use of the superficial temporal artery as a control artery for comparison with aneurysms limit the power of this study. Because of the difficulty in obtaining intracranial arteries in humans, use of the superficial temporal artery as a control, although not ideal, is warranted (17). However, extracranial arteries, specimens from autopsies, and other arteries harvested from the consequences of the temporal or frontal lobectomy can not be ideal controls for aneurysms. The development of a good experimental system could overcome the scarcity of clinical specimens and control arteries in understanding the pathogenesis of IA. Nevertheless, our data suggest that activation of TNFα, and its downstream signaling components, in aneurysms may constitute an important component of a proinflammatory mechanism that contributes to the pathological mechanisms of this disease. Our results set the basis for further studies on this multifunctional proinflammatory protein with regard to establishing new therapies for, and prevention of, human aneurysms.
Activation of the proinflammatory cytokine TNFα may contribute to the development of abnormal vasculature during aneurysm formation. The development of relevant experimental aneurysms that mimic important characteristics of human aneurysms will promote further studies toward understanding the contribution of proinflammatory and anti-inflammatory cytokines in aneurysm formation and rupture and will facilitate the development of effective methods to control these maladies.
This study was supported by a New Investigator Development Award and a Grant-in-Aid from the American Heart Association (TJ) and the Vascular Biology Fund. We thank Peter Rappa for administrative help.
In this article, the authors present a small study on a limited number of samples of human intracranial aneurysms compared with superficial temporal arteries. When analyzed for expression of tumor necrosis factor-α, this substance was found to be increased in the aneurysms. There was a lack of expression of interleukin-10 and an increase in fas-associated death domain (FADD) expression. The studies concluded that cytokine-related inflammation was upregulated in aneurysms. In addition, there was a lack of ability to regulate cytokines and turn them off, meaning that the process may be related to apoptosis in cell wall components. These results are intriguing and indeed would be of great importance. The study, however, is limited by its small sample size and a lack of quantification. Given the importance of the results, a statistical analysis would be appreciated. In addition, questions must be raised as to the suitability of the superficial temporal artery, which is an extracranial artery not related to the intracranial milieu, as a control. This is a common problem in aneurysmal wall studies that has never really been well satisfied because of the difficulty of obtaining human tissue. The findings themselves are very preliminary, and the conclusions assume a great deal about formative processes on the basis of these two findings. One must also state that it is not clear whether these aneurysms were ruptured or asymptomatic, because the process of previous bleed or even surgical delivery of this may stimulate inflammatory processes acutely, which are not part of the pathophysiology of the development of the aneurysm itself. Further review and statistical analysis would greatly strengthen the article.
Robert J. Dempsey
In this report, Jayaraman et al. demonstrated that expression of tumor necrosis factor-α and FADD protein was increased in 10 aneurysm specimens relative to control superficial temporal artery. These findings suggest that inflammation and apoptosis may participate in aneurysm formation and growth. The absence of interleukin-10 expression in aneurysms suggests that the responsible mediators of inflammation are activated monocytes/macrophages and T lymphocytes. It may be that sick patients with ruptured aneurysms have a stimulated inflammatory state and that elevated expression of tumor necrosis factor-α and FADD represents an epiphenomenon. However, the results of this study fit with an emerging understanding of aneurysm formation as a process of inflammatory degradation of vascular tissues. A similar process is observed in arteriovenous malformations that have hemorrhaged (1,2). These studies are of interest because they hint at new therapies for unruptured or untreatable aneurysms and arteriovenous malformations. Further studies from this group of researchers will be eagerly anticipated.
Michael T. Lawton
San Francisco, California
This is an important finding of overexpression of messenger ribonucleic acid of tumor necrosis factor-α and its downstream target FADD, but not interleukin-10, in human cerebral aneurysms in comparison with superficial temporal arteries. The authors discuss the implications in terms of potential mechanisms of inflammatory response in aneurysm wall. These preliminary results will need confirmation in a larger sample size and validation by protein expression studies or other complementary techniques. The results do not address whether this is a nonspecific “vascular-injury” response or a specific pathobiological manifestation of aneurysm disease. Other controls are needed (vascular malformation or atheromatous lesion). And the use of gene arrays (with similar quantities of messenger ribonucleic acid), including newer “custom” cytokine and inflammatory gene chips, allows the exploration of the whole inflammatory cascade, instead of selected molecules. The polymerase chain reaction can still be used to verify the results of key differentially expressed genes.
These results come on the heels of another important article, recently published by Frosen et al. (1), showing greater macrophage infiltration in the walls of ruptured than unruptured aneurysms, including lesions clipped within hours of hemorrhage, indicating inflammatory cell infiltration of aneurysm wall before rupture. The authors are encouraged to explore this inflammatory cascade in unruptured as well as ruptured aneurysms and in larger and smaller lesions. The molecular signatures of aneurysm growth and rupture, including the role of the inflammatory cascade, might open new doors for molecular imaging surveillance and therapeutic modification of aneurysm disease.
Issam A. Awad
The present study analyzed messenger ribonucleic acid expressions of three molecules: tumor necrosis factor-α, interleukin-10, and FADD, in human cerebral aneurysms, and superficial temporal arteries were taken as a control. Although they provided new data for particular inflammatory molecules in cerebral aneurysms, several issues should be resolved before the roles of these molecules are proved in the pathogenesis of cerebral aneurysms. Their data should be reinforced by measuring the bioactivity of each molecule. Small sample sizes also weakened the power of their data. Their data might be influenced by the condition of the aneurysms, such as ruptured or unruptured, size, neck, or dome. Their polymerase chain reaction data should be carefully interpreted because of quantitative accuracy and statistical weakness. Although the data provided in the present study cannot indicate the essential role of cytokines in the development of cerebral aneurysms, molecular analyses may give us a clue to new options in the treatment.