Matrix Metalloproteinase-8 and -9 Are Increased at the Site of Abdomin
http://www.100md.com
《循环学杂志》
the Departments of Surgery (W.R.W.W., M.A., E.C.S., P.R.F.B.) and Pathology (J.L.J., P.N.F.), University of Leicester, Leicester, UK, and Department of Vascular Surgery (M.M.T.), St George’s Hospital Medical School, London, UK.
Abstract
Background— Abdominal aortic aneurysm (AAA) expansion is characterized by extracellular matrix degradation and widespread inflammation. In contrast, the processes that characterize AAA rupture are not well understood. The aim of this study was to investigate the proteolytic and cellular activity of ruptured AAA, focusing on matrix metalloproteinases (MMPs) and their inhibitors (TIMPs).
Methods and Results— Anterior aneurysm wall biopsies were taken from 55 nonruptured and 21 ruptured AAAs. A further biopsy from the site of rupture was taken from 12 of the ruptured AAAs. MMP-1, -2, -3, -8, -9, and -13, as well as TIMP-1 and -2, were quantified in each biopsy with ELISA. A comparison of anterior aneurysm biopsies showed no difference in MMP or TIMP concentrations between nonruptured and ruptured AAA. In a comparison of ruptured AAA biopsies, MMP-8 and -9 levels were significantly elevated in the 12 rupture site biopsies compared with their 12 paired anterior wall biopsies, whereas other MMPs and TIMPs showed no difference (MMP-8, P<0.001; MMP-9, P=0.01). MMP-8 and -9 expression was mediated by native mesenchymal cells and was independent of the inflammatory infiltrate.
Conclusions— A localized increase in MMP-8 and –9, mediated by native mesenchymal cells, presents a potential pathway for collagen breakdown and AAA rupture.
Key Words: aneurysm inflammation metalloproteinases rupture
Introduction
The formation and expansion of an abdominal aortic aneurysm (AAA) are characterized by extracellular matrix degradation, increased proteolytic activity, and an inflammatory cell infiltrate.1,2 Fragmentation of elastin fibers and a reduction in elastin concentration appear to be mediated by increased concentrations of matrix metalloproteinases (MMPs) secreted by various cell types within the aortic wall.3 With progressive loss of elastin, established AAAs are composed largely of collagens types I and III.4
Clinical Perspective p 445
It seems reasonable to suggest that the final common pathway leading to aortic rupture might involve proteolytic degradation of the collagen matrix. This concept was first investigated by Dobrin et al,5 who investigated the proteolytic effects of purified collagenase and elastase on isolated arterial tissue perfused at supraphysiological pressures. Treatment with elastase caused the vessels to dilate markedly and become less compliant but was not related to rupture. In contrast, treatment with collagenase caused the blood vessels to dilate, become more compliant, and rupture. These findings supported the hypothesis that elastin degradation was a key step in aneurysmal dilatation but that collagen degradation was ultimately required for AAA rupture.
The role of MMPs in initiating AAA formation has been extensively delineated6 compared with only a limited number of studies examining MMP changes associated with continued AAA expansion and rupture.7–10 The aim of this study was to investigate the role of the potential collagenolytic MMPs in AAA rupture.
Methods
Study Design
Local research ethics committee approval for the study was obtained, and written consent was obtained from all patients. Two patient groups were studied: 55 patients with nonruptured AAAs undergoing elective repair and 21 patients with ruptured AAAs undergoing emergency surgery. The maximum external diameter of each nonruptured AAA was determined from a preoperative CT. The diameter of each ruptured AAA was measured intraoperatively.
Patient demographic information included age, gender, smoking history (current or ex-smoker of <10 years versus nonsmoker or ex-smoker of >10 years), presence of a cardiovascular event (documented myocardial infarction, cerebrovascular or peripheral vascular disease, angina requiring medication), hypertension (requiring medication), and diabetes (requiring medication or dietary modification). Cardiovascular medication was also recorded (statin, -blocker, calcium channel blocker, acetylcholine esterase inhibitor, and nonsteroidal antiinflammatory).
Tissue Collection
Aortic wall biopsies were obtained intraoperatively from the anterior aneurysm wall 5 cm distal to the left renal vein in all nonruptured and ruptured AAAs. When identified, an additional biopsy from each ruptured AAA was taken from the edge of the rupture site in 12 patients. Of the 12 identified rupture sites, 10 were posterior and 2 were anterior. No rupture site coincided with the site of anterior aneurysm wall biopsy. All biopsies were divided in 2, fixed in 4% paraformaldehyde, embedded in paraffin wax or washed in saline, dissected of luminal thrombus and adipose tissue, and frozen in liquid nitrogen.
MMP Quantification
MMPs and their endogenous tissue inhibitors (TIMPs) were extracted from tissues by an established methodology.11 In brief, frozen aortic biopsies were thawed and then homogenized in a volume of buffer solution directly proportional to the wet weight of the aortic biopsy. After centrifugation (12 000 rpm for 60 minutes at 4°C), the supernatant containing soluble protein extract was dialyzed (overnight at 4°C). The concentration of each protein extract was then standardized to 1 mg/mL by spectrophotometry.
ELISA kits (Amersham Pharmacia Biotech) were used to quantify the following MMPs from each protein extract: MMP-1, MMP-2, MMP-3, MMP-13, TIMP-1, TIMP-2 (total [t] levels for all), MMP-8, and MMP-9 (total [t] and active fraction [a] for both). The final concentration of each MMP and TIMP was expressed as nanograms of target protein per milligram of protein extract.
The MMP-8 and MMP-9 activity ELISAs use the proform of a detection enzyme, which undergoes proteolytic activation by the active form of MMP-8 or MMP-9. The degree of detection enzyme activation is then quantified with a chromogenic peptide substrate. Total MMP-8 and MMP-9 levels are measured by the same assay after the conversion of inactive MMP to the active form by the action of aminophenylmercuric acid. Assays measuring only total MMP or TIMP levels use conventional sandwich ELISA formats.
Histological Analysis
Anterior aneurysm wall biopsies from nonruptured (n=55) and ruptured (n=21) AAAs were analyzed, along with site of rupture biopsies (n=12). Five histological sections from each biopsy were stained with hematoxylin and eosin. Sections were examined for the presence of red blood cell contamination and evaluated by an investigator (JLJ) blinded to the site of each biopsy site.
MMP Immunohistochemistry and Colocalization Studies
Immunohistochemistry (IHC) for MMP-8 and MMP-9 was performed on a representative group of paired biopsies from ruptured AAAs. Histological sections were dewaxed in xylene and rehydrated through graded alcohols to water. Antigen retrieval was achieved with 0.1% trypsin solution in 0.1% calcium chloride, buffered to pH 7.8 (37°C for 12 minutes). After addition of the primary antibody for 1 hour at room temperature (either monoclonal MMP-8 [1:50] R&D Systems or MMP-9 [1:100] Novocastra Laboratories), the EnVision Detection System (EnVision, Dako) was used according the manufacturers instructions with washes between each step with PBS. Diaminobenzidine (Dako) at a concentration of 0.05% was added to each slide (room temperature for 10 minutes). The sections were then washed in distilled water, counterstained with hematoxylin for 15 seconds, and washed in tap water. Finally, the sections were dehydrated in absolute alcohol and mounted from clean xylene in mounting medium. Microscopically, areas of immunoreactivity were visualized as dark brown staining, demonstrating the diaminobenzidine being oxidized.
Colocalization studies for mesenchymal cells (anti-human vimentin [1:100], Dako), smooth muscle cells (SMCs; anti-human smooth muscle -actin [-SMA; 1:1000], Dako), T and B lymphocytes (anti-human CD-45, leukocyte common antigen [1:50], Dako), macrophages (anti-human CD-68 [1:150], Dako), and neutrophils (anti-human CD-15 [1:50], Becton Dickinson) were conducted. Negative controls involved exception of the primary antibody stage.
Morphometric Quantification of Cellular Content
Further IHC for mesenchymal cells, SMCs, lymphocytes, macrophages, and neutrophils was conducted on sections of nonruptured and ruptured AAA biopsies as detailed above. Quantification of the IHC staining was undertaken with a Nikon E800 microscope with an attached JVC KYF50 3 chip color video camera linked to an Apple G3 computer through a Scion CG-7 frame grabber. Images were analyzed with the freeware package NIH Image automated by in-house macros. In each case, the aortic adventitia was distinguished. Background illumination was digitally subtracted from each image before thresholding at a preset level, which distinguished consistently between stained and nonstained areas. A percentage area fraction of immunostaining per field area was calculated for each aneurysm biopsy from the mean of 10 images taken over 5 stained sections.
MMP In Situ Hybridization
Nonisotopic in situ hybridization (ISH) was performed for MMP-8 and MMP-9 mRNA on a representative group of paired biopsies from ruptured AAAs. Digoxigenin-labeled oligonucleotide cocktails were based on published sequences: MMP-812: 5'-T C G A C A G T C T C C G A C T C C A T C T T T C T C G A T-3'; 5'-C G G A A C G A C A G A G G G T T G A T A C G A A A G T C C-3'; 5'-T T G T A T G A A G A A A C A T T T A C T G G T T A A G A C-3'; and 5'-T C T T G A T C T A A A A C C A A T C T T C A T T C C T A A-3'; MMP-913: 5'-A C T G G C A G G G T T T C C C A T C A G C A T T G C C G T-3' , 5'-T C C G G C A C T G A G G A A T G A T C T A A G C C C A G C-3', 5'-G T T G C A G G C A T C G T C C A C C G G A C T C A A A G G-3', and 5'-G C T C C C C C T G C C C T C A G A G A A T C G C C A G T A-3'.
Histological sections were pretreated with proteinase K solution (4 μg/mL, 37°C for 20 minutes). The sections were incubated (65°C for 15 minutes, 37°C for 2 hours) in a hybridization solution (30% formamide, 0.6 mol/L NaCl, 10% dextran sulfate, 50 mmol/L Tris, pH 7.5, 0.1% sodium pyrophosphate, 0.2% Ficoll, and 5 mmol/L EDTA) containing either MMP-8 or MMP-9 oligonucleotide cocktails (or the respective sense equivalent as a control). Stringency washes were carried out with 2x standard citrate saline/30% formamide (37°C for 10 minutes). The sections were incubated with filtered bovine serum solution (10 minutes). Oligonucleotide detection used 100 μL antidigoxigenin alkaline phosphates (1 hour), followed by 5-bromo-4-chloro-3-indylphosphate and nitroblue tetrazolium (12 hours).
Statistical Analysis
Statistical analysis used GraphPad Prism 5. Discrete variables were presented as numbers and percentages and compared by use of Fisher’s exact test. The continuous variable age was normally distributed, presented as a mean (and SD), and compared by use of the independent t test. Other continuous variables were nonnormally distributed, reported as a median and interquartile range (AAA diameter, MMP levels, TIMP levels, and cellular quantification data), and compared by use of the Mann-Whitney U test or Wilcoxon paired test. Spearman’s correlation was used to test for correlations. Statistical significance was assumed at the P<0.01 level.
Results
Patient Demographics
The clinical features of the nonruptured and ruptured AAA patient cohorts are described in Table 1. Median AAA diameter was greater in ruptured than nonruptured AAA. There were no other differences in the characteristics of the study cohorts.
MMP and TIMP Concentrations in Nonruptured AAA: Correlation With Size
All the MMPs and TIMPs studied were detected by ELISA in the anterior aneurysm wall of nonruptured AAAs. There were no significant correlations between AAA diameter and enzyme concentrations (Table 2). These data suggested that there was no increase in the proteolytic capacity of the anterior aneurysm wall as the aneurysm sac increased in diameter. The ruptured AAA cohort was not included in this analysis because of difficulties in accurately determining diameter.
MMP and TIMP Concentrations in Anterior Aneurysm Wall Biopsies From Nonruptured and Ruptured AAAs
There were no differences in the median levels of the MMPs and TIMPs within anterior aneurysm wall biopsies from ruptured and nonruptured AAA (Table 3). These data strongly suggested that aneurysm rupture was not associated with a global change in MMP concentration throughout the aneurysm sac. Therefore, further analysis was performed to investigate any local differences within ruptured AAAs comparing the anterior aneurysm wall and the site of rupture.
MMP and TIMP Concentrations in Anterior Aneurysm Wall and Site of Aortic Rupture Biopsies From Ruptured AAAs
Table 4 details the paired biopsies from the anterior aneurysm wall and edge of rupture site from ruptured AAAs. There were no statistical differences in the concentrations of MMP-1(t), MMP-2(t), MMP-3(t), and MMP-13(t) or the inhibitors TIMP-1(t) and TIMP–2(t) between the paired rupture biopsies.
In contrast, the concentrations of total MMP-8 and MMP-9 were significantly elevated in biopsies from the site of aortic rupture compared with the paired biopsies of the anterior aneurysm wall: MMP-8(t), 43.6 ng/mg (range, 20.1 to 79.3 ng/mg) versus 17.0 ng/mg (range, 14.2 to 21.7 ng/mg) (P<0.001); MMP-9(t), 87.1 ng/mg (range, 56.3 to 113 ng/mg) versus 21.2 ng/mg (range, 7.79 to 75.5 ng/mg) (P=0.010). Furthermore, the active-fraction concentration of MMP-8 and -9 was significantly higher at the site of rupture: MMP-8(a), 10.3 ng/mg (range, 3.98 to 16.0 ng/mg) versus 2.80 ng/mg (range, 2.26 to 7.75 ng/mg) (P=0.005); MMP-9(a), 12.2 ng/mg (range, 7.77 to 16.2 ng/mg) versus 6.24 ng/mg (range, 4.43 to 15.5 ng/mg) (P=0.008). These differences suggested that there were selectively increased concentrations of 2 MMPs at the actual site of aortic rupture, which implicated these enzymes in the rupture process.
Histological Analysis
All biopsies from the anterior aneurysm wall and rupture site demonstrated destruction of the intima and media and prominence of vasa vasorum, which may represent neovascularization. Examination of all nonruptured and ruptured AAA biopsies showed no red blood cell contamination (Figures 1a, 2a, 3a, and 4a).
IHC and Colocalization Studies
IHC for MMP-8 and MMP-9 was performed on paired biopsies from ruptured AAAs. MMP-8 and MMP-9 were present at both the anterior aneurysm wall and rupture site biopsies with prominent adventitial staining for both MMPs (Figures 1b, 2b, 3b, and 4b). -SMA and vimentin-antigen immunostaining was noted throughout the anterior aneurysm wall and rupture site biopsies. A significant colocalization for -SMA and vimentin with MMP-8 and MMP-9 was observed (circled areas in Figure 1 through Figure 4). The expression of CD-45 and CD-68 was observed principally around the vasa vasorum (Figures 1f, 2f, 3f, and 4f and Figures 1e, 2e, 3e, and 4 e, respectively). Although present, IHC expression of CD-15 was low in all histological sections (Figures 1g, 2g, and 3g), with some sections demonstrating no neutrophils (Figure 4g).
Overall, the colocalization of -SMA to MMP-8 and -9 was slightly less consistent than that of vimentin, with occasional areas of vimentin-positive but -SMA–negative MMP staining observed. -SMA denotes SMCs; however, myofibroblasts and endothelial cells may also express -SMA.14 The vimentin antigen is present on mesenchymal cells, including fibroblasts and SMCs. The close localization of MMP-8 and MMP-9 staining with -SMA and vimentin and the failure of CD-68, CD-45, and CD-15 to colocalize, indicates the cellular source of these MMPs in the human aorta to be the native mesenchymal cells—fibroblasts, SMCs, or a myofibroblast intermediate.
Morphometric Quantification of Cellular Content
Table 5 details the quantitative morphological data taken from the anterior aneurysm wall biopsies of nonruptured and ruptured AAA. Overall, there was no difference in any observed parameter, with the percentage area fraction of immunostaining for CD-15, CD-45, CD-68, -SMA, and vimentin comparable between AAA groups.
Table 6 compares the quantitative morphological data from ruptured AAAs at the anterior aneurysm wall and the site of rupture. The percentage area fraction of CD-15, CD-68, -SMA, and vimentin failed to demonstrate significant differences between biopsy sites. Of note, however, the percentage area fraction of CD-45 was lower at the rupture site compared with paired anterior aneurysm wall biopsies (3.23% [range, 1.47% to 7.30%] versus 6.38% [range, 3.56% to 13.9%]; P=0.005).
In Situ Hybridization
ISH performed on paired rupture biopsies demonstrated MMP-8 and MMP-9 mRNA within the AAA wall at both the anterior aneurysm wall and the rupture site (Figures 1i, 2i, 3i, and 4i), supporting the native mesenchymal cell as the primary source of these 2 MMPs within the aneurysm wall.
Controls
Negative IHC controls confirmed the absence of nonspecific immunostaining in aneurysm biopsies (Figures 1h, 2h, 3h, and 4h). Negative ISH controls using complementary sense oligonucleotide sequences returned no positivity (Figures 1j, 2j, 3j, and 4j).
Discussion
The structural integrity of the aorta depends on elastin and fibrillar collagen types I and III.14,15 The in vitro model of Dobrin et al5 suggests that loss of elastin is responsible for early aortic expansion, whereas late expansion and rupture are modulated by collagen breakdown. The principal group of endogenous proteases implicated in aneurysm pathobiology are the MMPs. Traditionally, MMPs are subgrouped according to their substrate specificity: elastases MMP-2 and -9; collagenases MMP-1, -8, and -13; and stromelysin MMP-3. This classification is somewhat arbitrary because MMP-2 and -9 are now recognized to possess activity against partially degraded type I and III collagen16,17 and MMP-3 to activate tissue procollagenases.18
The elevation of MMP-2 and MMP-9 at protein level is well described in AAAs,7,10 and their critical role in AAA formation is supported by gene knockout studies.6 Similarly, elevation of MMP-1, MMP-8, and MMP-13 is described in AAA.1,19,20 Increased MMP-3 is reported at the transcription level,21 but protein concentrations of MMP-3 have not previously been described in aortic tissue. Reporting of TIMP levels in AAAs compared with normal controls is inconsistent, with some authors suggesting elevated and others observing decreased TIMP expression.20,22
The biochemical and cellular events associated with rupture of an abdominal aortic aneurysm have not been studied extensively. Hypothetically, 2 mechanisms might be considered to contribute to AAA rupture, namely a global or a local change in proteolytic activity. Because the incidence of aneurysm rupture increases as aortic diameter increases, it might be expected that global proteolytic capacity might increase as the aneurysm diameter increases, leading to an eventual threshold at which the capacity of the aortic wall to resist proteolytic degradation is exceeded and rupture ensues. Alternatively, the global proteolytic capacity of the aneurysm may remain unchanged, with an increased, intense, localized proteolysis in vulnerable portions of the aneurysm wall leading to localized collagen degradation and rupture.
A global elevation in MMP concentration has not been convincingly demonstrated in aneurysms of progressing diameter. Freestone et al7 described more prominent MMP-9 immunostaining in AAAs >5.5 cm in contrast to AAAs of a smaller diameter but observed a reciprocal result with MMP-2. Sakalihasan et al10 revealed that activated MMP-9 was present in AAAs with a mean diameter of 7.5 cm but was absent in AAAs with mean diameter of 5.8 cm. Similarly, McMillan et al8 demonstrated higher transcription levels of MMP-9 in medium-sized aneurysms (diameter, 5 to 6.9 cm) compared with both smaller and larger aneurysms.8 These data were interpreted to suggest that increased MMP-9 expression was related to the continued expansion of moderate-sized aneurysms, but the lower levels of MMP-9 expression in aneurysms >7 cm implied that the high rupture rates in large aneurysms were related to other factors. The present study did not demonstrate any positive relationship between AAA diameter and MMP concentration, which would suggest that there is no global elevation in proteolysis with increasing aneurysm size.
The current investigation also revealed that there was no difference in MMP concentrations between the anterior aneurysm wall of ruptured and nonruptured aneurysms. These data provide further evidence that a global upregulation of protease activity is not responsible for aneurysm rupture. Similar findings were reported by Peterson et al,9 who compared protein levels of MMP-9 and MMP-2 in stable and ruptured aneurysms. Unfortunately, the biopsy sites were not described in detail, but the findings did not suggest any clear difference in MMP-2 or -9 between these 2 groups overall.
If global changes in the aneurysm wall do not lead to rupture, it is possible that rupture may result from a localized process of elevated stress forces, resulting in accelerated matrix degradation. Vallabhaneni et al,23 reported clear differences in the tensile properties of nonaneurysmal aorta in the longitudinal and transverse directions and suggested that such anisotropy exists in aneurysms. A marked heterogeneity and high intersubject variation in aneurysm wall strength suggested that there were focal areas of weakened aortic wall, pointing to localized "hot spots" of MMP hyperactivity.23 Fillinger et al24 reported the study of 103 patients with initial aneurysm wall stress determined by finite-element analysis. Follow-up for 1 year, with 42 noninterventions, 39 elective repairs, and 22 emergency repairs, showed that initial peak wall stress correlated with the subsequent site of aneurysm rupture.24 Furthermore, Vorp et al25 found that wall stress increased substantially within asymmetric bulges of the aneurysm wall.
Data from the present investigations have revealed that there was a localized increase in MMP-8 and MMP-9 concentrations at the site of aortic rupture. Histological studies have demonstrated that these elevated enzyme levels were not associated with exogenous blood contamination or inflammatory infiltrate. The presence of MMP-9 is important not only because it appears to influence all stages of AAA pathogenesis but also because its substrate specificity toward partially degraded fibrillar collagen fragments17 complements the action of MMP-8, a potent type I collagenase.26 Given the predominance of type I collagen over type III in the aneurysmal aorta,4 the elevation of MMP-8 at the site of rupture is indeed significant and supports the original data of Dobrin et al.5 MMP-8 transcription was considered to occur only in maturing bone marrow–associated neutrophils.27 This assumption was rejected with the finding of MMP-8 transcripts in articular chondrocytes, synovial fibroblasts,28 and subsequently vascular-associated endothelial cells, macrophages, and SMCs.12 The presence of MMP-8 in the anterior aneurysm wall and rupture site, however, conflicts with data from Carrell et al,21 who were not able to detect MMP-8 transcripts in 8 aneurysm biopsies. The high levels of MMP-8 protein in the present study may reflect the fact that MMP-8 is stored as preformed protein granules and thus mRNA levels may not be representative of protein concentration.19
The cellular changes at the site of aneurysm rupture have not previously been delineated. The results of the IHC indicate that expression of MMP-8 and -9 was dependent on mesenchymal cells native to the aorta—either SMCs, fibroblasts, or myofibroblasts—rather than infiltrating inflammatory cells. Previous studies support the mesenchymal cell as a source of MMP-1, -2, -8, and -13.19,20,29,30 IHC studies localize MMP-9 to macrophages and native mesenchymal cells.29,31,32 Animal models present compelling evidence supporting the macrophage as a source of aortic MMP-9,6 but the findings from this study contradict this and demonstrates the limitations of relating animal models to human pathology. Quantification of the percentage area fraction immunostaining for CD-15 and CD-68 failed to demonstrate a change at the rupture site, whereas CD-45 was lower. Initially, this seems contradictory but suggests that inflammation has little bearing on the terminal process of aneurysm rupture. Indeed, the paucity of neutrophils observed in this study concurs with Cohen et al,33 who reported the absence of neutrophils except for an occasional neutrophil caught within the vasa vasorum.
A limitation of this study was the reliance on end-stage disease tissue. However, the site of anterior aneurysm wall biopsy, 5 cm distal to the left renal vein rather than at the point of maximum dilatation, was considered to represent the zone of proximal transition where the process of aneurysmal degeneration was likely to be the most aggressive.34 Use of paired aortic biopsies to illustrate an elevated MMP concentration at the site of aortic rupture negates many of the disadvantages of using clinical material. The difference in AAA size observed between the study groups did not bias the direct comparison of MMP concentrations between groups because MMP concentrations did not correlate with the diameter of nonruptured AAA. We recognize that the extraction of matrix-bound MMPs and TIMPs is reported to be variable29; however, our method used a well-established detergent-based protocol.11 Finally this study is relatively small, and it is possible that the statistical tests used were of low power.
These findings demonstrate for the first time that elevated MMP activity mediated by native mesenchymal cells may contribute to the rupture of an AAA. Further investigations clearly are needed to determine factors precipitating MMP expression and the role of inflammation in the pathogenesis of AAA rupture.
Acknowledgments
Disclosures
None.
References
Annabi B, Shedid D, Ghosn P, Kenigsberg RL, Desrosiers RR, Bojanowski MW, Beaulieu E, Nassif E, Moumdjian R, Beliveau R. Differential regulation of matrix metalloproteinase activities in abdominal aortic aneurysms. J Vasc Surg. 2002; 35: 539–546.
Holmes DR, Liao S, Parks WC, Thompson RW. Medial neovascularization in abdominal aortic aneurysms: a histopathologic marker of aneurysmal degeneration with pathophysiologic implications. J Vasc Surg. 1995; 21: 761–771.
Sakalihasan N, Heyeres A, Nusgens BV, Limet R, Lapiere CM. Modifications of the extracellular matrix of aneurysmal abdominal aortas as a function of their size. Eur J Vasc Surg. 1993; 7: 633–637.
Rizzo RJ, McCarthy WJ, Dixit SN, Lilly MP, Shively VP, Flinn WR, Yao JS. Collagen types and matrix protein content in human abdominal aortic aneurysms. J Vasc Surg. 1989; 10: 365–373.
Dobrin PB, Baker WH, Gley WC. Elastolytic and collagenolytic studies of arteries: implications for the mechanical properties of aneurysms. Arch Surg. 1984; 119: 405–409.
Longo GM, Xiong W, Greiner TC, Zhao Y, Fiotti N, Baxter BT. Matrix metalloproteinases 2 and 9 work in concert to produce aortic aneurysm. J Clin Invest. 2002; 110: 625–632.
Freestone T, Turner RJ, Coady A, Higman DJ, Greenhalgh RM, Powell JT. Inflammation and matrix metalloproteinases in the enlarging abdominal aortic aneurysm. Arterioscler Thromb Vasc Biol. 1995; 15: 1145–1151.
McMillan WD, Tamarina NA, Cipollone M, Johnson DA, Parker MA, Pearce WH. Size matters: the relationship between MMP-9 expression and aortic diameter. Circulation. 1997; 96: 2228–2232.
Petersen E, Wagberg F, Angquist KA. Proteolysis of the abdominal aortic aneurysm wall and the association with rupture. Eur J Vasc Endovasc Surg. 2002; 23: 153–157.
Sakalihasan N, Delvenne P, Nusgens BV, Limet R, Lapiere CM. Activated forms of MMP2 and MMP9 in abdominal aortic aneurysms. J Vasc Surg. 1996; 24: 127–133.
Vine N, Powell JT. Metalloproteinases in degenerative aortic disease. Clin Sci. 1991; 81: 233–239.
Herman MP, Sukhova GK, Libby P, Gerdes N, Tang N, Horton DB, Kilbride M, Breitbart RE, Chun M, Schonbeck U. Expression of neutrophil collagenase (matrix metalloproteinase-8) in human atheroma: a novel collagenolytic pathway suggested by transcriptional profiling. Circulation. 2001; 104: 1899–1904.
Loftus IM, Naylor AR, Goodall S, Crowther M, Jones L, Bell PR, Thompson MM. Increased matrix metalloproteinase-9 activity in unstable carotid plaques: a potential role in acute plaque disruption. Stroke. 2000; 31: 40–47.
Kazi M, Thyberg J, Religa P, Roy J, Eriksson P, Hedin U, Swedenborg J. Influence of intraluminal thrombus on structural and cellular composition of abdominal aortic aneurysm wall. J Vasc Surg. 2003; 38: 1283–1292.
Dobrin PB, Mrkvicka R. Failure of elastin or collagen as possible critical connective tissue alterations underlying aneurysmal dilatation. Cardiovasc Surg. 1994; 2: 484–488.
Aimes RT, Quigley JP. Matrix metalloproteinase-2 is an interstitial collagenase: inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. J Biol Chem. 1995; 270: 5872–5876.
Watanabe H, Nakanishi I, Yamashita K, Hayakawa T, Okada Y. Matrix metalloproteinase-9 (92 kDa gelatinase/type IV collagenase) from U937 monoblastoid cells: correlation with cellular invasion. J Cell Sci. 1993; 104 (pt 4): 991–999.
Suzuki K, Enghild T, Morodomi G, Salvesen G, Hagase H. The activation of tissue procollagenase by matrix metalloproteinase 3 (stromelysin). Biochemistry. 1990; 29: 10261–10270.
Mao D, Lee JK, VanVickle SJ, Thompson RW. Expression of collagenase-3 (MMP-13) in human abdominal aortic aneurysms and vascular smooth muscle cells in culture. Biochem Biophy Res Commun. 1999; 261: 904–910.
Wilson WRW, Schwable EC, Jones JL, Bell PRF, Thompson MM. MMP-8 (neutrophil collagenase): the missing collagenase in abdominal aortic aneurysm pathogenesis. Br J Surg. 2005; 92: 828–833.
Carrell TW, Burnand KG, Wells GM, Clements JM, Smith A. Stromelysin-1 (matrix metalloproteinase-3) and tissue inhibitor of metalloproteinase-3 are overexpressed in the wall of abdominal aortic aneurysms. Circulation. 2002; 105: 477–482.
Brophy CM, Reilly JM, Smith GJ, Tilson MD. The role of inflammation in nonspecific abdominal aortic aneurysm disease. Ann Vasc Surg. 1991; 5: 229–233.
Vallabhaneni SR, Gilling-Smith GL, How TV, Carter SD, Brennan JA, Harris PL. Heterogeneity of tensile strength and matrix metalloproteinase activity in the wall of abdominal aortic aneurysms. J Endovasc Ther. 2004; 11: 494–502.
Fillinger MF, Marra SP, Raghavan ML, Kennedy FE. Prediction of rupture risk in abdominal aortic aneurysm during observation: wall stress versus diameter. J Vasc Surg. 2003; 37: 724–732.
Vorp DA, Raghavan ML, Webster MW. Mechanical wall stress in abdominal aortic aneurysm: influence of diameter and asymmetry. J Vasc Surg. 1998; 27: 632–639.
Horwitz AL, Hance AJ, Crystal RG. Granulocyte collagenase: selective digestion of type I relative to type III collagen. Proc Natl Acad Sci U S A. 1977; 74: 897–901.
Hasty KA, Jeffrey JJ, Hibbs MS, Welgus HG. The collagen substrate specificity of human neutrophil collagenase. J Biol Chem. 1987; 262: 10048–10052.
Cole AA, Chubinskaya S, Schumacher B, Huch K, Szabo G, Yao J, Mikecz K, Hasty KA, Kuettner KE. Chondrocyte matrix metalloproteinase-8: human articular chondrocytes express neutrophil collagenase. J Biol Chem. 1996; 271: 11023–11026.
Davis V, Persidskaia R, Baca-Regen L, Itoh Y, Nagase H, Persidsky Y, Ghorpade A, Baxter BT. Matrix metalloproteinase-2 production and its binding to the matrix are increased in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 1998; 18: 1625–1633.
Sasaguri Y, Murahashi N, Sugama K, Kato S, Hiraoka K, Satoh T, Isomoto H, Morimatsu M. Development-related changes in matrix metalloproteinase expression in human aortic smooth muscle cells. Lab Invest. 1994; 71: 261–269.
Irizarry E, Newman KM, Gandhi RH, Nackman GB, Halpern V, Wishner S, Scholes JV, Tilson MD. Demonstration of interstitial collagenase in abdominal aortic aneurysm disease. J Surg Res. 1993; 54: 571–574.
Patel MI, Melrose J, Ghosh P, Appleberg M. Increased synthesis of matrix metalloproteinases by aortic smooth muscle cells is implicated in the etiopathogenesis of abdominal aortic aneurysms. J Vasc Surg. 1996; 24: 82–92.
Cohen JR, Keegan L, Sarfati I, Danna D, Ilardi C, Wise L. Neutrophil chemotaxis and neutrophil elastase in the aortic wall in patients with abdominal aortic aneurysms. J Invest Surg. 1991; 4: 423–430.
Curci JA, Liao S, Huffman MD, Shapiro SD, Thompson RW. Expression and localization of macrophage elastase (matrix metalloproteinase-12) in abdominal aortic aneurysms. J Clin Invest. 1998; 102: 1900–1910.(W. Richard W. Wilson, BSc, MRCS; Marcus )
Abstract
Background— Abdominal aortic aneurysm (AAA) expansion is characterized by extracellular matrix degradation and widespread inflammation. In contrast, the processes that characterize AAA rupture are not well understood. The aim of this study was to investigate the proteolytic and cellular activity of ruptured AAA, focusing on matrix metalloproteinases (MMPs) and their inhibitors (TIMPs).
Methods and Results— Anterior aneurysm wall biopsies were taken from 55 nonruptured and 21 ruptured AAAs. A further biopsy from the site of rupture was taken from 12 of the ruptured AAAs. MMP-1, -2, -3, -8, -9, and -13, as well as TIMP-1 and -2, were quantified in each biopsy with ELISA. A comparison of anterior aneurysm biopsies showed no difference in MMP or TIMP concentrations between nonruptured and ruptured AAA. In a comparison of ruptured AAA biopsies, MMP-8 and -9 levels were significantly elevated in the 12 rupture site biopsies compared with their 12 paired anterior wall biopsies, whereas other MMPs and TIMPs showed no difference (MMP-8, P<0.001; MMP-9, P=0.01). MMP-8 and -9 expression was mediated by native mesenchymal cells and was independent of the inflammatory infiltrate.
Conclusions— A localized increase in MMP-8 and –9, mediated by native mesenchymal cells, presents a potential pathway for collagen breakdown and AAA rupture.
Key Words: aneurysm inflammation metalloproteinases rupture
Introduction
The formation and expansion of an abdominal aortic aneurysm (AAA) are characterized by extracellular matrix degradation, increased proteolytic activity, and an inflammatory cell infiltrate.1,2 Fragmentation of elastin fibers and a reduction in elastin concentration appear to be mediated by increased concentrations of matrix metalloproteinases (MMPs) secreted by various cell types within the aortic wall.3 With progressive loss of elastin, established AAAs are composed largely of collagens types I and III.4
Clinical Perspective p 445
It seems reasonable to suggest that the final common pathway leading to aortic rupture might involve proteolytic degradation of the collagen matrix. This concept was first investigated by Dobrin et al,5 who investigated the proteolytic effects of purified collagenase and elastase on isolated arterial tissue perfused at supraphysiological pressures. Treatment with elastase caused the vessels to dilate markedly and become less compliant but was not related to rupture. In contrast, treatment with collagenase caused the blood vessels to dilate, become more compliant, and rupture. These findings supported the hypothesis that elastin degradation was a key step in aneurysmal dilatation but that collagen degradation was ultimately required for AAA rupture.
The role of MMPs in initiating AAA formation has been extensively delineated6 compared with only a limited number of studies examining MMP changes associated with continued AAA expansion and rupture.7–10 The aim of this study was to investigate the role of the potential collagenolytic MMPs in AAA rupture.
Methods
Study Design
Local research ethics committee approval for the study was obtained, and written consent was obtained from all patients. Two patient groups were studied: 55 patients with nonruptured AAAs undergoing elective repair and 21 patients with ruptured AAAs undergoing emergency surgery. The maximum external diameter of each nonruptured AAA was determined from a preoperative CT. The diameter of each ruptured AAA was measured intraoperatively.
Patient demographic information included age, gender, smoking history (current or ex-smoker of <10 years versus nonsmoker or ex-smoker of >10 years), presence of a cardiovascular event (documented myocardial infarction, cerebrovascular or peripheral vascular disease, angina requiring medication), hypertension (requiring medication), and diabetes (requiring medication or dietary modification). Cardiovascular medication was also recorded (statin, -blocker, calcium channel blocker, acetylcholine esterase inhibitor, and nonsteroidal antiinflammatory).
Tissue Collection
Aortic wall biopsies were obtained intraoperatively from the anterior aneurysm wall 5 cm distal to the left renal vein in all nonruptured and ruptured AAAs. When identified, an additional biopsy from each ruptured AAA was taken from the edge of the rupture site in 12 patients. Of the 12 identified rupture sites, 10 were posterior and 2 were anterior. No rupture site coincided with the site of anterior aneurysm wall biopsy. All biopsies were divided in 2, fixed in 4% paraformaldehyde, embedded in paraffin wax or washed in saline, dissected of luminal thrombus and adipose tissue, and frozen in liquid nitrogen.
MMP Quantification
MMPs and their endogenous tissue inhibitors (TIMPs) were extracted from tissues by an established methodology.11 In brief, frozen aortic biopsies were thawed and then homogenized in a volume of buffer solution directly proportional to the wet weight of the aortic biopsy. After centrifugation (12 000 rpm for 60 minutes at 4°C), the supernatant containing soluble protein extract was dialyzed (overnight at 4°C). The concentration of each protein extract was then standardized to 1 mg/mL by spectrophotometry.
ELISA kits (Amersham Pharmacia Biotech) were used to quantify the following MMPs from each protein extract: MMP-1, MMP-2, MMP-3, MMP-13, TIMP-1, TIMP-2 (total [t] levels for all), MMP-8, and MMP-9 (total [t] and active fraction [a] for both). The final concentration of each MMP and TIMP was expressed as nanograms of target protein per milligram of protein extract.
The MMP-8 and MMP-9 activity ELISAs use the proform of a detection enzyme, which undergoes proteolytic activation by the active form of MMP-8 or MMP-9. The degree of detection enzyme activation is then quantified with a chromogenic peptide substrate. Total MMP-8 and MMP-9 levels are measured by the same assay after the conversion of inactive MMP to the active form by the action of aminophenylmercuric acid. Assays measuring only total MMP or TIMP levels use conventional sandwich ELISA formats.
Histological Analysis
Anterior aneurysm wall biopsies from nonruptured (n=55) and ruptured (n=21) AAAs were analyzed, along with site of rupture biopsies (n=12). Five histological sections from each biopsy were stained with hematoxylin and eosin. Sections were examined for the presence of red blood cell contamination and evaluated by an investigator (JLJ) blinded to the site of each biopsy site.
MMP Immunohistochemistry and Colocalization Studies
Immunohistochemistry (IHC) for MMP-8 and MMP-9 was performed on a representative group of paired biopsies from ruptured AAAs. Histological sections were dewaxed in xylene and rehydrated through graded alcohols to water. Antigen retrieval was achieved with 0.1% trypsin solution in 0.1% calcium chloride, buffered to pH 7.8 (37°C for 12 minutes). After addition of the primary antibody for 1 hour at room temperature (either monoclonal MMP-8 [1:50] R&D Systems or MMP-9 [1:100] Novocastra Laboratories), the EnVision Detection System (EnVision, Dako) was used according the manufacturers instructions with washes between each step with PBS. Diaminobenzidine (Dako) at a concentration of 0.05% was added to each slide (room temperature for 10 minutes). The sections were then washed in distilled water, counterstained with hematoxylin for 15 seconds, and washed in tap water. Finally, the sections were dehydrated in absolute alcohol and mounted from clean xylene in mounting medium. Microscopically, areas of immunoreactivity were visualized as dark brown staining, demonstrating the diaminobenzidine being oxidized.
Colocalization studies for mesenchymal cells (anti-human vimentin [1:100], Dako), smooth muscle cells (SMCs; anti-human smooth muscle -actin [-SMA; 1:1000], Dako), T and B lymphocytes (anti-human CD-45, leukocyte common antigen [1:50], Dako), macrophages (anti-human CD-68 [1:150], Dako), and neutrophils (anti-human CD-15 [1:50], Becton Dickinson) were conducted. Negative controls involved exception of the primary antibody stage.
Morphometric Quantification of Cellular Content
Further IHC for mesenchymal cells, SMCs, lymphocytes, macrophages, and neutrophils was conducted on sections of nonruptured and ruptured AAA biopsies as detailed above. Quantification of the IHC staining was undertaken with a Nikon E800 microscope with an attached JVC KYF50 3 chip color video camera linked to an Apple G3 computer through a Scion CG-7 frame grabber. Images were analyzed with the freeware package NIH Image automated by in-house macros. In each case, the aortic adventitia was distinguished. Background illumination was digitally subtracted from each image before thresholding at a preset level, which distinguished consistently between stained and nonstained areas. A percentage area fraction of immunostaining per field area was calculated for each aneurysm biopsy from the mean of 10 images taken over 5 stained sections.
MMP In Situ Hybridization
Nonisotopic in situ hybridization (ISH) was performed for MMP-8 and MMP-9 mRNA on a representative group of paired biopsies from ruptured AAAs. Digoxigenin-labeled oligonucleotide cocktails were based on published sequences: MMP-812: 5'-T C G A C A G T C T C C G A C T C C A T C T T T C T C G A T-3'; 5'-C G G A A C G A C A G A G G G T T G A T A C G A A A G T C C-3'; 5'-T T G T A T G A A G A A A C A T T T A C T G G T T A A G A C-3'; and 5'-T C T T G A T C T A A A A C C A A T C T T C A T T C C T A A-3'; MMP-913: 5'-A C T G G C A G G G T T T C C C A T C A G C A T T G C C G T-3' , 5'-T C C G G C A C T G A G G A A T G A T C T A A G C C C A G C-3', 5'-G T T G C A G G C A T C G T C C A C C G G A C T C A A A G G-3', and 5'-G C T C C C C C T G C C C T C A G A G A A T C G C C A G T A-3'.
Histological sections were pretreated with proteinase K solution (4 μg/mL, 37°C for 20 minutes). The sections were incubated (65°C for 15 minutes, 37°C for 2 hours) in a hybridization solution (30% formamide, 0.6 mol/L NaCl, 10% dextran sulfate, 50 mmol/L Tris, pH 7.5, 0.1% sodium pyrophosphate, 0.2% Ficoll, and 5 mmol/L EDTA) containing either MMP-8 or MMP-9 oligonucleotide cocktails (or the respective sense equivalent as a control). Stringency washes were carried out with 2x standard citrate saline/30% formamide (37°C for 10 minutes). The sections were incubated with filtered bovine serum solution (10 minutes). Oligonucleotide detection used 100 μL antidigoxigenin alkaline phosphates (1 hour), followed by 5-bromo-4-chloro-3-indylphosphate and nitroblue tetrazolium (12 hours).
Statistical Analysis
Statistical analysis used GraphPad Prism 5. Discrete variables were presented as numbers and percentages and compared by use of Fisher’s exact test. The continuous variable age was normally distributed, presented as a mean (and SD), and compared by use of the independent t test. Other continuous variables were nonnormally distributed, reported as a median and interquartile range (AAA diameter, MMP levels, TIMP levels, and cellular quantification data), and compared by use of the Mann-Whitney U test or Wilcoxon paired test. Spearman’s correlation was used to test for correlations. Statistical significance was assumed at the P<0.01 level.
Results
Patient Demographics
The clinical features of the nonruptured and ruptured AAA patient cohorts are described in Table 1. Median AAA diameter was greater in ruptured than nonruptured AAA. There were no other differences in the characteristics of the study cohorts.
MMP and TIMP Concentrations in Nonruptured AAA: Correlation With Size
All the MMPs and TIMPs studied were detected by ELISA in the anterior aneurysm wall of nonruptured AAAs. There were no significant correlations between AAA diameter and enzyme concentrations (Table 2). These data suggested that there was no increase in the proteolytic capacity of the anterior aneurysm wall as the aneurysm sac increased in diameter. The ruptured AAA cohort was not included in this analysis because of difficulties in accurately determining diameter.
MMP and TIMP Concentrations in Anterior Aneurysm Wall Biopsies From Nonruptured and Ruptured AAAs
There were no differences in the median levels of the MMPs and TIMPs within anterior aneurysm wall biopsies from ruptured and nonruptured AAA (Table 3). These data strongly suggested that aneurysm rupture was not associated with a global change in MMP concentration throughout the aneurysm sac. Therefore, further analysis was performed to investigate any local differences within ruptured AAAs comparing the anterior aneurysm wall and the site of rupture.
MMP and TIMP Concentrations in Anterior Aneurysm Wall and Site of Aortic Rupture Biopsies From Ruptured AAAs
Table 4 details the paired biopsies from the anterior aneurysm wall and edge of rupture site from ruptured AAAs. There were no statistical differences in the concentrations of MMP-1(t), MMP-2(t), MMP-3(t), and MMP-13(t) or the inhibitors TIMP-1(t) and TIMP–2(t) between the paired rupture biopsies.
In contrast, the concentrations of total MMP-8 and MMP-9 were significantly elevated in biopsies from the site of aortic rupture compared with the paired biopsies of the anterior aneurysm wall: MMP-8(t), 43.6 ng/mg (range, 20.1 to 79.3 ng/mg) versus 17.0 ng/mg (range, 14.2 to 21.7 ng/mg) (P<0.001); MMP-9(t), 87.1 ng/mg (range, 56.3 to 113 ng/mg) versus 21.2 ng/mg (range, 7.79 to 75.5 ng/mg) (P=0.010). Furthermore, the active-fraction concentration of MMP-8 and -9 was significantly higher at the site of rupture: MMP-8(a), 10.3 ng/mg (range, 3.98 to 16.0 ng/mg) versus 2.80 ng/mg (range, 2.26 to 7.75 ng/mg) (P=0.005); MMP-9(a), 12.2 ng/mg (range, 7.77 to 16.2 ng/mg) versus 6.24 ng/mg (range, 4.43 to 15.5 ng/mg) (P=0.008). These differences suggested that there were selectively increased concentrations of 2 MMPs at the actual site of aortic rupture, which implicated these enzymes in the rupture process.
Histological Analysis
All biopsies from the anterior aneurysm wall and rupture site demonstrated destruction of the intima and media and prominence of vasa vasorum, which may represent neovascularization. Examination of all nonruptured and ruptured AAA biopsies showed no red blood cell contamination (Figures 1a, 2a, 3a, and 4a).
IHC and Colocalization Studies
IHC for MMP-8 and MMP-9 was performed on paired biopsies from ruptured AAAs. MMP-8 and MMP-9 were present at both the anterior aneurysm wall and rupture site biopsies with prominent adventitial staining for both MMPs (Figures 1b, 2b, 3b, and 4b). -SMA and vimentin-antigen immunostaining was noted throughout the anterior aneurysm wall and rupture site biopsies. A significant colocalization for -SMA and vimentin with MMP-8 and MMP-9 was observed (circled areas in Figure 1 through Figure 4). The expression of CD-45 and CD-68 was observed principally around the vasa vasorum (Figures 1f, 2f, 3f, and 4f and Figures 1e, 2e, 3e, and 4 e, respectively). Although present, IHC expression of CD-15 was low in all histological sections (Figures 1g, 2g, and 3g), with some sections demonstrating no neutrophils (Figure 4g).
Overall, the colocalization of -SMA to MMP-8 and -9 was slightly less consistent than that of vimentin, with occasional areas of vimentin-positive but -SMA–negative MMP staining observed. -SMA denotes SMCs; however, myofibroblasts and endothelial cells may also express -SMA.14 The vimentin antigen is present on mesenchymal cells, including fibroblasts and SMCs. The close localization of MMP-8 and MMP-9 staining with -SMA and vimentin and the failure of CD-68, CD-45, and CD-15 to colocalize, indicates the cellular source of these MMPs in the human aorta to be the native mesenchymal cells—fibroblasts, SMCs, or a myofibroblast intermediate.
Morphometric Quantification of Cellular Content
Table 5 details the quantitative morphological data taken from the anterior aneurysm wall biopsies of nonruptured and ruptured AAA. Overall, there was no difference in any observed parameter, with the percentage area fraction of immunostaining for CD-15, CD-45, CD-68, -SMA, and vimentin comparable between AAA groups.
Table 6 compares the quantitative morphological data from ruptured AAAs at the anterior aneurysm wall and the site of rupture. The percentage area fraction of CD-15, CD-68, -SMA, and vimentin failed to demonstrate significant differences between biopsy sites. Of note, however, the percentage area fraction of CD-45 was lower at the rupture site compared with paired anterior aneurysm wall biopsies (3.23% [range, 1.47% to 7.30%] versus 6.38% [range, 3.56% to 13.9%]; P=0.005).
In Situ Hybridization
ISH performed on paired rupture biopsies demonstrated MMP-8 and MMP-9 mRNA within the AAA wall at both the anterior aneurysm wall and the rupture site (Figures 1i, 2i, 3i, and 4i), supporting the native mesenchymal cell as the primary source of these 2 MMPs within the aneurysm wall.
Controls
Negative IHC controls confirmed the absence of nonspecific immunostaining in aneurysm biopsies (Figures 1h, 2h, 3h, and 4h). Negative ISH controls using complementary sense oligonucleotide sequences returned no positivity (Figures 1j, 2j, 3j, and 4j).
Discussion
The structural integrity of the aorta depends on elastin and fibrillar collagen types I and III.14,15 The in vitro model of Dobrin et al5 suggests that loss of elastin is responsible for early aortic expansion, whereas late expansion and rupture are modulated by collagen breakdown. The principal group of endogenous proteases implicated in aneurysm pathobiology are the MMPs. Traditionally, MMPs are subgrouped according to their substrate specificity: elastases MMP-2 and -9; collagenases MMP-1, -8, and -13; and stromelysin MMP-3. This classification is somewhat arbitrary because MMP-2 and -9 are now recognized to possess activity against partially degraded type I and III collagen16,17 and MMP-3 to activate tissue procollagenases.18
The elevation of MMP-2 and MMP-9 at protein level is well described in AAAs,7,10 and their critical role in AAA formation is supported by gene knockout studies.6 Similarly, elevation of MMP-1, MMP-8, and MMP-13 is described in AAA.1,19,20 Increased MMP-3 is reported at the transcription level,21 but protein concentrations of MMP-3 have not previously been described in aortic tissue. Reporting of TIMP levels in AAAs compared with normal controls is inconsistent, with some authors suggesting elevated and others observing decreased TIMP expression.20,22
The biochemical and cellular events associated with rupture of an abdominal aortic aneurysm have not been studied extensively. Hypothetically, 2 mechanisms might be considered to contribute to AAA rupture, namely a global or a local change in proteolytic activity. Because the incidence of aneurysm rupture increases as aortic diameter increases, it might be expected that global proteolytic capacity might increase as the aneurysm diameter increases, leading to an eventual threshold at which the capacity of the aortic wall to resist proteolytic degradation is exceeded and rupture ensues. Alternatively, the global proteolytic capacity of the aneurysm may remain unchanged, with an increased, intense, localized proteolysis in vulnerable portions of the aneurysm wall leading to localized collagen degradation and rupture.
A global elevation in MMP concentration has not been convincingly demonstrated in aneurysms of progressing diameter. Freestone et al7 described more prominent MMP-9 immunostaining in AAAs >5.5 cm in contrast to AAAs of a smaller diameter but observed a reciprocal result with MMP-2. Sakalihasan et al10 revealed that activated MMP-9 was present in AAAs with a mean diameter of 7.5 cm but was absent in AAAs with mean diameter of 5.8 cm. Similarly, McMillan et al8 demonstrated higher transcription levels of MMP-9 in medium-sized aneurysms (diameter, 5 to 6.9 cm) compared with both smaller and larger aneurysms.8 These data were interpreted to suggest that increased MMP-9 expression was related to the continued expansion of moderate-sized aneurysms, but the lower levels of MMP-9 expression in aneurysms >7 cm implied that the high rupture rates in large aneurysms were related to other factors. The present study did not demonstrate any positive relationship between AAA diameter and MMP concentration, which would suggest that there is no global elevation in proteolysis with increasing aneurysm size.
The current investigation also revealed that there was no difference in MMP concentrations between the anterior aneurysm wall of ruptured and nonruptured aneurysms. These data provide further evidence that a global upregulation of protease activity is not responsible for aneurysm rupture. Similar findings were reported by Peterson et al,9 who compared protein levels of MMP-9 and MMP-2 in stable and ruptured aneurysms. Unfortunately, the biopsy sites were not described in detail, but the findings did not suggest any clear difference in MMP-2 or -9 between these 2 groups overall.
If global changes in the aneurysm wall do not lead to rupture, it is possible that rupture may result from a localized process of elevated stress forces, resulting in accelerated matrix degradation. Vallabhaneni et al,23 reported clear differences in the tensile properties of nonaneurysmal aorta in the longitudinal and transverse directions and suggested that such anisotropy exists in aneurysms. A marked heterogeneity and high intersubject variation in aneurysm wall strength suggested that there were focal areas of weakened aortic wall, pointing to localized "hot spots" of MMP hyperactivity.23 Fillinger et al24 reported the study of 103 patients with initial aneurysm wall stress determined by finite-element analysis. Follow-up for 1 year, with 42 noninterventions, 39 elective repairs, and 22 emergency repairs, showed that initial peak wall stress correlated with the subsequent site of aneurysm rupture.24 Furthermore, Vorp et al25 found that wall stress increased substantially within asymmetric bulges of the aneurysm wall.
Data from the present investigations have revealed that there was a localized increase in MMP-8 and MMP-9 concentrations at the site of aortic rupture. Histological studies have demonstrated that these elevated enzyme levels were not associated with exogenous blood contamination or inflammatory infiltrate. The presence of MMP-9 is important not only because it appears to influence all stages of AAA pathogenesis but also because its substrate specificity toward partially degraded fibrillar collagen fragments17 complements the action of MMP-8, a potent type I collagenase.26 Given the predominance of type I collagen over type III in the aneurysmal aorta,4 the elevation of MMP-8 at the site of rupture is indeed significant and supports the original data of Dobrin et al.5 MMP-8 transcription was considered to occur only in maturing bone marrow–associated neutrophils.27 This assumption was rejected with the finding of MMP-8 transcripts in articular chondrocytes, synovial fibroblasts,28 and subsequently vascular-associated endothelial cells, macrophages, and SMCs.12 The presence of MMP-8 in the anterior aneurysm wall and rupture site, however, conflicts with data from Carrell et al,21 who were not able to detect MMP-8 transcripts in 8 aneurysm biopsies. The high levels of MMP-8 protein in the present study may reflect the fact that MMP-8 is stored as preformed protein granules and thus mRNA levels may not be representative of protein concentration.19
The cellular changes at the site of aneurysm rupture have not previously been delineated. The results of the IHC indicate that expression of MMP-8 and -9 was dependent on mesenchymal cells native to the aorta—either SMCs, fibroblasts, or myofibroblasts—rather than infiltrating inflammatory cells. Previous studies support the mesenchymal cell as a source of MMP-1, -2, -8, and -13.19,20,29,30 IHC studies localize MMP-9 to macrophages and native mesenchymal cells.29,31,32 Animal models present compelling evidence supporting the macrophage as a source of aortic MMP-9,6 but the findings from this study contradict this and demonstrates the limitations of relating animal models to human pathology. Quantification of the percentage area fraction immunostaining for CD-15 and CD-68 failed to demonstrate a change at the rupture site, whereas CD-45 was lower. Initially, this seems contradictory but suggests that inflammation has little bearing on the terminal process of aneurysm rupture. Indeed, the paucity of neutrophils observed in this study concurs with Cohen et al,33 who reported the absence of neutrophils except for an occasional neutrophil caught within the vasa vasorum.
A limitation of this study was the reliance on end-stage disease tissue. However, the site of anterior aneurysm wall biopsy, 5 cm distal to the left renal vein rather than at the point of maximum dilatation, was considered to represent the zone of proximal transition where the process of aneurysmal degeneration was likely to be the most aggressive.34 Use of paired aortic biopsies to illustrate an elevated MMP concentration at the site of aortic rupture negates many of the disadvantages of using clinical material. The difference in AAA size observed between the study groups did not bias the direct comparison of MMP concentrations between groups because MMP concentrations did not correlate with the diameter of nonruptured AAA. We recognize that the extraction of matrix-bound MMPs and TIMPs is reported to be variable29; however, our method used a well-established detergent-based protocol.11 Finally this study is relatively small, and it is possible that the statistical tests used were of low power.
These findings demonstrate for the first time that elevated MMP activity mediated by native mesenchymal cells may contribute to the rupture of an AAA. Further investigations clearly are needed to determine factors precipitating MMP expression and the role of inflammation in the pathogenesis of AAA rupture.
Acknowledgments
Disclosures
None.
References
Annabi B, Shedid D, Ghosn P, Kenigsberg RL, Desrosiers RR, Bojanowski MW, Beaulieu E, Nassif E, Moumdjian R, Beliveau R. Differential regulation of matrix metalloproteinase activities in abdominal aortic aneurysms. J Vasc Surg. 2002; 35: 539–546.
Holmes DR, Liao S, Parks WC, Thompson RW. Medial neovascularization in abdominal aortic aneurysms: a histopathologic marker of aneurysmal degeneration with pathophysiologic implications. J Vasc Surg. 1995; 21: 761–771.
Sakalihasan N, Heyeres A, Nusgens BV, Limet R, Lapiere CM. Modifications of the extracellular matrix of aneurysmal abdominal aortas as a function of their size. Eur J Vasc Surg. 1993; 7: 633–637.
Rizzo RJ, McCarthy WJ, Dixit SN, Lilly MP, Shively VP, Flinn WR, Yao JS. Collagen types and matrix protein content in human abdominal aortic aneurysms. J Vasc Surg. 1989; 10: 365–373.
Dobrin PB, Baker WH, Gley WC. Elastolytic and collagenolytic studies of arteries: implications for the mechanical properties of aneurysms. Arch Surg. 1984; 119: 405–409.
Longo GM, Xiong W, Greiner TC, Zhao Y, Fiotti N, Baxter BT. Matrix metalloproteinases 2 and 9 work in concert to produce aortic aneurysm. J Clin Invest. 2002; 110: 625–632.
Freestone T, Turner RJ, Coady A, Higman DJ, Greenhalgh RM, Powell JT. Inflammation and matrix metalloproteinases in the enlarging abdominal aortic aneurysm. Arterioscler Thromb Vasc Biol. 1995; 15: 1145–1151.
McMillan WD, Tamarina NA, Cipollone M, Johnson DA, Parker MA, Pearce WH. Size matters: the relationship between MMP-9 expression and aortic diameter. Circulation. 1997; 96: 2228–2232.
Petersen E, Wagberg F, Angquist KA. Proteolysis of the abdominal aortic aneurysm wall and the association with rupture. Eur J Vasc Endovasc Surg. 2002; 23: 153–157.
Sakalihasan N, Delvenne P, Nusgens BV, Limet R, Lapiere CM. Activated forms of MMP2 and MMP9 in abdominal aortic aneurysms. J Vasc Surg. 1996; 24: 127–133.
Vine N, Powell JT. Metalloproteinases in degenerative aortic disease. Clin Sci. 1991; 81: 233–239.
Herman MP, Sukhova GK, Libby P, Gerdes N, Tang N, Horton DB, Kilbride M, Breitbart RE, Chun M, Schonbeck U. Expression of neutrophil collagenase (matrix metalloproteinase-8) in human atheroma: a novel collagenolytic pathway suggested by transcriptional profiling. Circulation. 2001; 104: 1899–1904.
Loftus IM, Naylor AR, Goodall S, Crowther M, Jones L, Bell PR, Thompson MM. Increased matrix metalloproteinase-9 activity in unstable carotid plaques: a potential role in acute plaque disruption. Stroke. 2000; 31: 40–47.
Kazi M, Thyberg J, Religa P, Roy J, Eriksson P, Hedin U, Swedenborg J. Influence of intraluminal thrombus on structural and cellular composition of abdominal aortic aneurysm wall. J Vasc Surg. 2003; 38: 1283–1292.
Dobrin PB, Mrkvicka R. Failure of elastin or collagen as possible critical connective tissue alterations underlying aneurysmal dilatation. Cardiovasc Surg. 1994; 2: 484–488.
Aimes RT, Quigley JP. Matrix metalloproteinase-2 is an interstitial collagenase: inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. J Biol Chem. 1995; 270: 5872–5876.
Watanabe H, Nakanishi I, Yamashita K, Hayakawa T, Okada Y. Matrix metalloproteinase-9 (92 kDa gelatinase/type IV collagenase) from U937 monoblastoid cells: correlation with cellular invasion. J Cell Sci. 1993; 104 (pt 4): 991–999.
Suzuki K, Enghild T, Morodomi G, Salvesen G, Hagase H. The activation of tissue procollagenase by matrix metalloproteinase 3 (stromelysin). Biochemistry. 1990; 29: 10261–10270.
Mao D, Lee JK, VanVickle SJ, Thompson RW. Expression of collagenase-3 (MMP-13) in human abdominal aortic aneurysms and vascular smooth muscle cells in culture. Biochem Biophy Res Commun. 1999; 261: 904–910.
Wilson WRW, Schwable EC, Jones JL, Bell PRF, Thompson MM. MMP-8 (neutrophil collagenase): the missing collagenase in abdominal aortic aneurysm pathogenesis. Br J Surg. 2005; 92: 828–833.
Carrell TW, Burnand KG, Wells GM, Clements JM, Smith A. Stromelysin-1 (matrix metalloproteinase-3) and tissue inhibitor of metalloproteinase-3 are overexpressed in the wall of abdominal aortic aneurysms. Circulation. 2002; 105: 477–482.
Brophy CM, Reilly JM, Smith GJ, Tilson MD. The role of inflammation in nonspecific abdominal aortic aneurysm disease. Ann Vasc Surg. 1991; 5: 229–233.
Vallabhaneni SR, Gilling-Smith GL, How TV, Carter SD, Brennan JA, Harris PL. Heterogeneity of tensile strength and matrix metalloproteinase activity in the wall of abdominal aortic aneurysms. J Endovasc Ther. 2004; 11: 494–502.
Fillinger MF, Marra SP, Raghavan ML, Kennedy FE. Prediction of rupture risk in abdominal aortic aneurysm during observation: wall stress versus diameter. J Vasc Surg. 2003; 37: 724–732.
Vorp DA, Raghavan ML, Webster MW. Mechanical wall stress in abdominal aortic aneurysm: influence of diameter and asymmetry. J Vasc Surg. 1998; 27: 632–639.
Horwitz AL, Hance AJ, Crystal RG. Granulocyte collagenase: selective digestion of type I relative to type III collagen. Proc Natl Acad Sci U S A. 1977; 74: 897–901.
Hasty KA, Jeffrey JJ, Hibbs MS, Welgus HG. The collagen substrate specificity of human neutrophil collagenase. J Biol Chem. 1987; 262: 10048–10052.
Cole AA, Chubinskaya S, Schumacher B, Huch K, Szabo G, Yao J, Mikecz K, Hasty KA, Kuettner KE. Chondrocyte matrix metalloproteinase-8: human articular chondrocytes express neutrophil collagenase. J Biol Chem. 1996; 271: 11023–11026.
Davis V, Persidskaia R, Baca-Regen L, Itoh Y, Nagase H, Persidsky Y, Ghorpade A, Baxter BT. Matrix metalloproteinase-2 production and its binding to the matrix are increased in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 1998; 18: 1625–1633.
Sasaguri Y, Murahashi N, Sugama K, Kato S, Hiraoka K, Satoh T, Isomoto H, Morimatsu M. Development-related changes in matrix metalloproteinase expression in human aortic smooth muscle cells. Lab Invest. 1994; 71: 261–269.
Irizarry E, Newman KM, Gandhi RH, Nackman GB, Halpern V, Wishner S, Scholes JV, Tilson MD. Demonstration of interstitial collagenase in abdominal aortic aneurysm disease. J Surg Res. 1993; 54: 571–574.
Patel MI, Melrose J, Ghosh P, Appleberg M. Increased synthesis of matrix metalloproteinases by aortic smooth muscle cells is implicated in the etiopathogenesis of abdominal aortic aneurysms. J Vasc Surg. 1996; 24: 82–92.
Cohen JR, Keegan L, Sarfati I, Danna D, Ilardi C, Wise L. Neutrophil chemotaxis and neutrophil elastase in the aortic wall in patients with abdominal aortic aneurysms. J Invest Surg. 1991; 4: 423–430.
Curci JA, Liao S, Huffman MD, Shapiro SD, Thompson RW. Expression and localization of macrophage elastase (matrix metalloproteinase-12) in abdominal aortic aneurysms. J Clin Invest. 1998; 102: 1900–1910.(W. Richard W. Wilson, BSc, MRCS; Marcus )