Challenges and opportunities in abdominal aortic aneurysm research.

Abdominal Aortic Aneurysms (AAAs) are associated with advanced age, male gender, cigarette smoking, atherosclerosis, hypertension, and genetic predisposition. Basic research studies have led to a better understanding of aneurysm disease over the past two decades. There has also been a growing appreciation that fundamental knowledge regarding the process of aneurysmal degeneration is still somewhat limited. Opportunities in research include: 1) the investigation of potential new mechanism-based pharmacologic interventions; 2) identify the genetic basis for an inherited predisposition; 3) develop and refine noninvasive approaches for the early detection; 4) examine potential novel surgical approaches and design new biomaterials; and 5) initiate and promote awareness programs for diagnosis and treatment of aortic aneurysms. The optimal approach to addressing these issues will require integrative, multidisciplinary research programs that involve basic scientists working in concert with vascular and cardiothoracic surgeons, as well as other clinical specialists with expertise in vascular disease.

Features and genomic origins of matrix cell adhesion molecules-1 and -2 expressed by fibroblasts of human aortic adventitial origin.

Precursor mRNAs for proteins that we have called matrix cell adhesion molecules-1 and -2 (Mat-CAM-1 and Mat-CAM-2) were cloned from fibroblasts cultured from a specimen of human abdominal aorta. Both protein sequences have the unusual feature that there is an immunoglobulin kappa (Igkappa)-like domain at the N terminus and they are glycine/proline rich in the C-terminal domain. Antibodies were raised in rabbit against peptides synthesized for a specific unique sequence of Mat-CAM-1 and Mat-CAM-2, respectively (not detected in any other proteins referenced in GenBank). Both antibodies were immunoreactive with adventitial microfibrils of the aorta and some additional arteries. Mat-CAM-1 was detected in the common/internal iliac arteries, but it was not detected in the external iliac artery. Conversely, Mat-CAM-2 was conspicuous in the external iliac artery, but not the common/internal iliac arteries. Thus, these proteins show features of site-specific expression within the arterial tree. Computerized searches show that the genomic origins of Mat-CAM-1 and -2 are in the so-called nasopharnygocarcinoma (NPC) gene on chromosome 2, which is a putative oncogene. Expression of the two different gene products from the same genomic sequence requires shifts of reading frame. Further studies will be required to determine whether these proteins are prototypes for a larger family of tissue-specific matrix proteins arising from alternative transcriptions of the same genomic sequence.

Arterial aneurysms in HIV patients: molecular mimicry versus direct infection?

There is growing literature on the subject of aneurysmal degeneration of arteries in patients who are infected with HIV. A patient recently seen at our medical center with an aneurysm of the carotid artery stimulated our interest in reviewing the mechanisms by which HIV may initiate or predispose to these pathologies. There are at least three major possibilities: (1) immunodeficiency allows bacteria that are known to cause mycotic aneurysms to proliferate without immune restraint; (2) one or more of the HIV envelope proteins sufficiently resemble one or more artery-specific-antigenic proteins (ASAPs) that may trigger an autoimmune response (molecular mimicry); and (3) the HIV virus itself infects arterial-resident cells that maintain the integrity of the load-bearing matrix. The computational searches reported here suggest that the ASAP, matrix cell adhesion molecule-1 (Mat-CAM-1), has a high degree of similarity to known ligands for HIV envelope proteins gp41 and gp120. No similarities of Mat-CAM-1 to the HIV envelope glycoproteins were detected. Accordingly, among the possibilities for explaining the HIV/aneurysm connection, direct infection of aortic fibroblasts by the HIV virus is more likely to be the pathogenetic mechanism than the process of molecular mimicry.

Two decades of research on etiology and genetic factors in the abdominal aortic aneurysm (AAA)--with a glimpse into the 21st century.

Long gone are the days when the question of "etiology" of the abdominal aortic aneurysm (AAA) required a simplistic response. AAA is caused by an interplay of environmental and genetic factors, each of which may modify the expression of the other. A low penetrance, an insidious onset, and a wide distribution have forced scientists to rely on complex approaches to elucidate the pathophysiology of AAA. Hypotheses over the last twenty years have evaluated several components of connective tissue structure, function, and regulation. Although there has been considerable overlap in the many genetic approaches undertaken to explain aneurysm development, research recently has focused on reverse-engineering techniques. While earlier investigations using a "forward" or "candidate gene approach" have provided many insights in the understanding of AAA disease, advances in statistical modeling, new techniques in molecular biology, and gross computing power have now made more feasible the development of a reverse approach. New hope lies with the development of biochemical and computing tools which have paralleled information's vast dissemination in this rapidly widening field. Aortic inflammation and the upregulation of proteases and dysfunctional tissue-turnover are receiving more attention. No doubt, the interplay of all of these technologies will occupy investigators with an interest in AAA etiology for many years to come.

The nitrite/collagen reaction: non-enzymatic nitration as a model system for age-related damage.

The effects of age seen in long-lived connective tissue proteins are thought to be the result of post-translational modifications by reactive molecules. One such molecule is the nitrite ion. Human nitrite exposure results predominately from endogenous production of nitric oxide as well as inhalation of cigarette smoke and ingestion of cured meats. Although nitrite reactions with various proteins have been studied previously with regard to carcinogenesis, the specific reaction with collagen and its role in age-related damage has never been examined. We describe the reaction of nitrite with type I collagen at neutral pH and body temperature. The incubation of collagen with nitrite results in an increase in cross-linking, the accumulation of a yellow chromophore, and a depletion of tyrosine residues. Similar changes also are found in aged human collagen. In addition, 3-nitro-tyrosine, which has recently been used as a marker for peroxynitrite mediated damage, is produced from this reaction. Thus, we propose non-enzymatic nitration as an in vitro model system for human collagen age-related damage.

Abdominal aortic aneurysm as an autoimmune disease.

The pathogenesis of abdominal aortic aneurysms (AAAs) is unknown. We hypothesize that the autoimmune disease process plays a key role in the development of AAAs. Both cellular and humoral immunity is involved in the pathogenesis of AAAs. Triggers of autoimmunity are multifactorial. Certain HLA typing is closely related to AAAs, and a certain viral infection may have a potential role in the etiology of AAA via a molecular mimicry mechanism. The autoantigen is located in the microfibrillar compartment of the aortic wall as a normal structure. Patients with AAA are immunoreactive with this novel structural protein. If in the future the autoantigen is fully elucidated, serum testing to detect antibody against the autoantigen can be performed.

Successful repair of myocardial free wall rupture after thrombolytic therapy for acute infarction.

BACKGROUND: Controversy exists regarding the timing of thrombolytic administration and rupture rate. METHODS: Hospital records at St. Luke's-Roosevelt Hospital of the 4 study patients were reviewed and compared with those of 41 patients from a group of 537 patients concurrently admitted with a diagnosis of myocardial infarction (MI). RESULTS: Four patients experienced ventricular free wall rupture after having a MI between November 17, 1993, and July 28, 1995. All received tissue plasminogen activator. In 1 patient, pericardial effusion associated with a pseudoaneurysm was discovered in the operating room. The 3 others developed clinical pericardial tamponade before surgery. All 4 patients survived and left the hospital on postoperative days 10, 11, 11, and 82, respectively. During this same time period, 537 patients were admitted with MI, 41 of whom died; the study's 4 patients were compared with these 41. CONCLUSIONS: These data demonstrate that rupture of the ventricular free wall can occur early after thrombolytic therapy and may have a subacute course. Prompt diagnosis and surgery offer excellent chances of surviving this fatal condition.

A novel hypothesis regarding the evolutionary origins of the immunoglobulin fold.

The recent cloning of two cDNAs (Clone 1 and Clone 5) that encode novel hypothetical proteins, combining an N-terminal Ig kappa-like domain with features that occur in microfibril-associated glycoproteins (MAGPs) and fibrinogen, raises the question of whether the Ig fold may have originated in association with functions that may be more primitive than soluble immunity. Pairwise alignments were performed to compare similarities of fibrinogen-beta, Clone 1 and an Ig kappa sequence. Clone 1 had two regions in its Ig domain with > 50% similarity to fibrinogen, while Ig kappa was virtually non-homologous to fibrinogen. This result suggests that Clone 1 is closer to their common ancestor. A neighbour-joining tree was computed, and it supported this interpretation. Three-dimensional modelling of the most highly conserved sequence revealed two antiparallel beta strands connected by a helix. These observations suggest that the ancestral gene for the immunoglobulin superfamily may have originated as a primitive sandwich-like fold, possibly used in matrix/cell communications.

Expression of molecular messages for angiogenesis by fibroblasts from aneurysmal abdominal aorta versus dermal fibroblasts.

BACKGROUND: The molecular messages which drive angiogenesis in the adventitia of an aneurysmal aorta are uncertain. The emergence of molecular phenotyping by cDNA expression arrays provides a simple and rapid method for a preliminary approach to the analysis of molecular messengers for neovascularization in cultured cells. In the present experiment, fibroblasts cultured from the aorta of a patient with abdominal aortic aneurysm (AAA) were compared with normal dermal fibroblasts, on an array that evaluates several mRNAs with known roles in angiogenesis. METHODS: RNA was isolated from fibroblasts and purified. Labelled cDNA probes were generated from a mixture of RNA and CDS primers. Atlas Array membranes (Clontech) were prehybridized by ExpressHyb buffer. The cDNA probes were then added to the membranes, which were exposed to Phospholmage Screen (Molecular Dynamics) and analysed by a dedicated computer program. RESULTS: The most significantly upregulated mRNAs in AAA (by comparison to dermal fibroblasts) were: MCAF, MDNCF, EGR-1, VEGF, FGF-7, Mal protein, Mac Marcks, Transducin, Interleukin-9 receptor, and TNF. CONCLUSION: VEGF and TNF were upregulated, as expected. However, the upregulation of monocyte chemotactic and activating factor (MCAF) and monocyte-derived neutrophil chemotactic factor (MDNCF) suggest that the fibroblast may be more significantly involved in driving the inflammatory response that leads to AAA than previously realized.

Regional distribution in the mouse of proteins homologous to artery-specific antigenic proteins (ASAPs).

We have reported the sequences of four novel proteins derived from extracts of human aortic tissue and a cDNA library from human aortic adventitia. These proteins are immunoreactive with serum immunoglobulins from patients with abdominal aortic aneurysms (AAA), and they have homologies of amino acid sequence with microfibrillar proteins of connective tissue. We are reporting separately that two of these proteins are artery-specific antigenic proteins (ASAPs) in man. The present work investigates the regional distribution of these two proteins (AAAP-40 and MatCAM-1) in mouse (E-beta-b). Antibodies were raised in rabbit against polypeptides encoding novel amino acid sequences, unique to these proteins (e.g., not reported in GenBank). Immunohistochemical studies with these two specific antibodies show conspicuous immunoreactivity of collagen-associated microfibrils in the aortic adventitia of the murine abdominal and thoracic aorta. Immunoreactive peptides were not present in brain, muscle, or kidney. These findings support the hypothesis that proteins occur in the mouse that are homologous to a unique family of aortic microfibrillar proteins in man.

Regional distribution in human of a novel aortic collagen-associated microfibrillar protein.

We have reported two hypothetical proteins of human aorta, based on sequences cloned from a cDNA library constructed from mRNAs purified from the adventitia. These sequences have immunoglobulin-kappa (IgK)-like domains, and we have shown that microfibrils of the aortic adventia are immunoreactive with antibodies against IgK. The present study was performed to characterize more specifically the regional distribution in human of one of these proteins in particular and the distribution of matrix proteins with IgK-like motifs in general. An antibody was raised in rabbit against a synthetic peptide based on a unique sequence in one of the hypothetical proteins (NPSNRVTPQKNFP), which has not been reported in the sequence of any other known protein. This antibody and a rabbit anti-human IgK antibody were used as first antibodies in the staining reactions. A monoclonal mouse anti-smooth muscle cell alpha-actin antibody was also used. Immunoreactivity with the sequence-specific antibody was limited to the aorta and large vessels. Adventitial microfibrillar staining was more conspicuous in the abdominal aorta than in the thoracic aorta and in the internal iliac than in the external iliac artery. The immunoreactive protein was associated with fibroblasts and not smooth muscle cells. Immunoreactivity coated the collagen fibers diffusely, while elastin fibers were not stained. Further studies using antibodies against IgK demonstrated immunoreactivity of collagen and fibroblasts in a variety of tissues: spleen, ovary, testicle, cervix, prostate, skin, and breast (but not brain). Immunoglobulin motifs may be a feature of matrix proteins produced by fibroblasts in many tissues, but the first motif that we identified from a cDNA library of aortic adventitia appears to be specific to aorta and large vessels.

Immunoreactivity of adventitial matrix fibrils of normal and aneurysmal abdominal aorta with antibodies against vitronectin and fibrinogen.

PURPOSE: We have reported that immunoglobulin G (IgG) harvested from specimens of aneurysmal abdominal aorta (AAA) is immunoreactive with a fibrillar component of the matrix of the aortic adventitia. In further studies we have reported the partial amino acid sequence of a 40-kDa protein, purified from the adventitia of the human aorta, which we have called aortic aneurysm antigenic protein-40 kDa (AAAP-40). AAAP-40 has homologies with bovine microfibril-associated glycoprotein-36 kDa (MAGP-36). Both AAAP-40 and MAGP-36 have homologies to fibrinogen beta (FB-b) and vitronectin (VN). The purposes of the present experiments were (1) to determine whether antibodies against VN and fibrinogen are immunoreactive with elements of the normal and/or aneurysmal aortic wall, and (2) to determine whether these antibodies are immunoreactive with soluble extracts of aortic proteins. METHODS: Paraffin-embedded tissue sections of normal and aneurysmal aorta were probed with polyclonal rabbit antihuman VN and antihuman FB-b antibodies. Immunoblots of soluble aortic proteins were evaluated with the same antibodies. Histochemical preparations with Gomori's aldehyde fuchsin and elastin-von-Gieson solutions were also performed. RESULTS: Anti-VN and FB-b antibodies reacted with matrix fibrils in the aortic adventitia in both normal and aneurysmal specimens, with a distribution that resembles the appearance of fibrils that are stained by Gomori's reaction. By comparison to normal adventitial fibrils, fibrils in the specimens from AAA appeared fragmented, coiled, and frayed. Antibodies against VN and FB-b were immunoreactive in immunoblots with a soluble aortic protein of molecular weight approximately 40 kDa, consistent with the migration of AAAP-40. CONCLUSIONS: There appear to be immunodeterminants in AAAP-40 that are recognized by polyclonal antibodies against VN and FB-b.

Genetic risk factor for abdominal aortic aneurysm: HLA-DR2(15), a Japanese study.

PURPOSE: Autoimmunity has been proposed to play a role in the pathogenesis of the abdominal aortic aneurysm (AAA). Several autoimmune diseases are associated with specific HLA DR alleles. These experiments were carried out to determine whether the same HLA DR types that have been reported to be associated with AAA in a mixed North American population are similarly associated with AAA in a more homogeneous group of patients in Japan. METHOD: HLA DR typing was performed by a serologic method on samples of peripheral blood of patients with nonspecific infrarenal AAA in Nagasaki University Hospital in Japan. The frequencies of HLA DR antigens were compared with those of volunteers approximately matched for age and sex from the same referral area. RESULTS: HLA DR haplotypes were determined in 46 Japanese patients with AAA and in 50 patients in a control group. The HLA-DR2(15) antigen was observed in 27 (58.7%) patients (29 alleles 31.5%) with AAA and in 14 (28%) subjects (16 alleles 26.0%) in the control group (p < 0.005). CONCLUSIONS: The data suggest that HLA-DR2(15) has an important role as a genetic risk factor for AAA in Japanese patients, as previously reported in a mixed North American population.

Molecular cloning of the complementary DNA for an additional member of the family of aortic aneurysm antigenic proteins.

PURPOSE: We have purified and partially sequenced a protein from the adventitia of the human aorta (aortic aneurysm antigenic protein 40 kDa; AAAP-40) that has homologies to bovine aortic microfibril-associated glycoprotein (MAGP-36). It is immunoreactive with immunoglobulin G (IgGs) purified from the serum and aortic wall of patients with abdominal aortic aneurysms. AAAP-40 and MAGP-36 have fibrinogen-like and vitronectin-like motifs. Screening an expression library constructed from human aortic adventitial messenger RNA has resulted in the cloning of three complementary DNAs whose gene products are immunoreactive with immunoglobulin G from patients with abdominal aortic aneurysms. Two strongly resemble each other and have been described separately. The purpose of this article is to report the third clone. METHODS: Messenger RNA from a specimen of human aortic aneurysmal adventitia was reverse-transcribed for insertion into the phagemid Uni Zap XR (Stratagene). A strain of Escherichia coli, engineered for expression (XL 1-Blue MFR', Stratagene), was transfected, and rabbit antihuman vitronectin antibody as used to identify positive clones. Sequencing of the positive clones was performed by the Core Laboratories at Columbia University. RESULTS: The hypothetical protein of rAAAP-CL4 (clone 4) shares sequence motifs with known microfibril-associated glycoproteins (MAGPs). The recombinant protein (rAAAP-CL4) is immunoreactive with serum from patients (three of four abdominal aortic aneurysm sera). In addition, similarities have been detected with immunoglobulins of the kappa family and with a protein from cytomegalovirus that is a potential molecular mimic. CONCLUSIONS: There may several members of a novel family of human aortic autoantigenic proteins implicated in abdominal aortic aneurysm disease.

Elastin degradation products induce adventitial angiogenesis in the Anidjar/Dobrin rat aneurysm model.

BACKGROUND: Infusion of the abdominal aorta with pancreatic elastase induces aneurysms in a rat model (Anidjar/Dobrin). Because elastolysis liberates elastin degradation products (EDPs), the present experiment was carried out to test the hypothesis that an EDP alone could induce features of aneurysm disease. METHODS: The EDP val/gly/val/ala/pro/gly (VGVAPG), elastase, or saline solution was infused into infrarenal aorta (n = 4/group). After 1 week aortic diameters were measured, and the tissues were prepared for histologic examination. Adventitial capillaries (vessels per high-power field) were counted over a standardized preparation of aorta. Wall thickness was measured by means of computer-aided planimetry. RESULTS: There was an increase of greater than 100-fold in mean vessels per high-power field in aortas receiving VGVAPG or elastase versus saline controls (4.10 +/- 0.68 SEM or 4.48 +/- 0.49 SEM versus 0.03 +/- 0.03 SEM, respectively, p < 0.05). The VGVAPG-perfused group had a 26% +/- 4% SEM increase in diameter from baseline that was statistically significant (p < 0.01), but the aortas did not reach aneurysmal dimensions. CONCLUSIONS: Although no aneurysms occurred at 1 week after the infusion of EDP, the results demonstrate that the EDP VGVAPG can induce a characteristic feature of aneurysm disease. The model permits study of the earliest stages of experimental aneurysm formation and raises interesting questions regarding the role of the vasa vasorum in this pathologic process.

Expression of two novel recombinant proteins from aortic adventitia (kappafibs) sharing amino acid sequences with cytomegalovirus.

We have recently purified and partially sequenced a microfibrillar protein from human aortic adventitia (aneurysm-associated antigenic protein, 40 kDa [AAAP-40]) that is immunoreactive with immunoglobulin (IgG) from the wall of abdominal aortic aneurysms (AAAs). It shares motifs with Ig kappa (which may act as a binding site for interaction with integrins), cytomegalovirus (which may be a molecular mimic in the pathogenesis of AAA), and vitronectin and the fibrinogens. A cDNA library was constructed from the aortic adventitia of a AAA. The library was screened with either rabbit anti-vitronectin antibody or rabbit anti-fibrinogen antibody. Positive plaques were purified and expressed in a strain of Escherichia coli. The clone sequences were analyzed. The expressed proteins were separated by SDS/PAGE and the immunoblots were probed with either AAA IgG or anti-human Ig kappa antibody. Experimental cell lines, transfected with the clones (clones 1 and 5), synthesized recombinant proteins (rAAAP-CL1 and rAAAP-CL5), detectable in Western immunoblots with AAA IgG. A prediction of the tertiary structure resembles well-characterized cell adhesion molecules. These findings suggest that there is a novel family of matrix proteins that may use immunoglobulin motifs as binding sites for cellular integrins and that there are matrix proteins in addition to AAAP-40 that may serve as autoantigens in the pathogenesis of AAA disease.