Dissecting abdominal aortic aneurysm in Ang II-infused mice: the importance of imaging Bram Trachet1,2, Rodrigo Fraga-Silva2, Alessandra Piersigilli3,4, Patrick Segers1, Nikolaos Stergiopulos2 1 IBiTech - bioMMeda, Ghent University-iMinds Medical IT, Ghent, Belgium 2 Institute of Bioengineering, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland 3 School of Life Sciences, PTEC GE, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland 4 Institute of Animal Pathology, University of Bern, Bern, Switzerland Corresponding Author: Bram Trachet LHTC STI IBI EPFL BM 5128 Station 17 CH-1015 Lausanne (Switzerland) Tel: +41 21 693 83 81 Fax: +41 21 693 96 35 [email protected] Abstract Introduction. Since the initial publication in 2000, Angiotensin II-infused mice have become the model of choice to study abdominal aortic aneurysm in a pre-clinical setting. We recently used phase contrast X-ray based computed tomography to challenge the existing paradigm and demonstrated that these animals develop an apparent luminal dilatation and an intramural hematoma, both related to mural ruptures in the tunica media in the vicinity of suprarenal side branches, providing evidence that the main lesion is rather a mere dissection without previous dilation. Aims. The aim of this narrative review was to provide an extensive overview of small animal applicable techniques that have provided relevant insight into the pathogenesis and morphology of dissecting AAA in mice, and to relate findings from these techniques to each other and to our recent PCXTM-based results. Combining insights from recent and consolidated publications we aimed to enhance our understanding of dissecting AAA morphology and anatomy. Results and conclusions. We analyzed in vivo and ex vivo images of aortas obtained from macroscopic anatomy, histology, high-frequency ultrasound, contrast-enhanced micro-CT, micro- MRI and PCXTM. We demonstrate how in almost all publications the aorta has been subdivided into a part in which an intact lumen lies adjacent to a remodeled wall/hematoma, and a part in which elastic lamellae are ruptured and the lumen appears to be dilated. We show how the novel paradigm fits within the existing one, and how 3D images can explain and connect previously published 2D structures that had never been correctly interpreted. We conclude that PCXTM- based findings are in line with previous results, and all evidence points towards the fact that dissecting AAAs in Angiotensin II-infused mice are actually caused by ruptures of the tunica media in the immediate vicinity of small side branches. Keywords: abdominal aortic aneurysms, small animal imaging, dissecting aneurysm, high- frequency ultrasound, PCXTM, micro-CT, MRI, angiotensin II, mouse model Introduction Abdominal aortic aneurysm (AAA) is defined as a local dilation of aortic diameter that exceeds 150% of the initial size (1). Since the first paper published by Daugherty et al in 2000 (2), the Angiotensin (Ang) II-infused mouse model for AAA formation has been the golden standard for pre-clinical aneurysm research, and was used in (roughly) 200 papers studying abdominal aortic aneurysm in a preclinical setting. The model has been reported to replicate many features of human aneurysm formation, such as (amongst others) elastin degradation, luminal dilatation, and thrombus formation. Nevertheless, it has long been unknown why the murine AAA occurs suprarenally rather than infrarenally, why there is a high variability in aneurysm shape, and why the animals develop intramural rather than intraluminal thrombus. Notwithstanding the fact that most – if not all – early publications (approximately up to 2010) focused on the many similarities between human and murine AAAs, we have recently published results that challenge the existing paradigm of Ang II-induced AAAs in mice. Using a novel imaging technique called PCXTM (phase contrast X-ray computed tomography) we demonstrated that these animals in fact develop dissecting AAAs, and present an intramural hematoma due to ruptures of the aortic tunica media in the vicinity of small supraceliac side branches. Moreover, we demonstrated that the apparent primary luminal dilatation in these dissecting AAAs is actually due to the mural dissection of the tunica media in the segment proximal to the celiac artery. At first sight, these observations contradicted a large number of previous studies, all of which have reported “true” luminal dilatation and none of which reported branch-related ruptures leading to intramural hematoma. Understanding and interpretation are however limited by how much can be directly observed on a given sample, hence also to the intrinsic properties of the imaging technique that is used. Despite the fact that preclinical imaging technologies are usually technically much more advanced than the clinical devices, they could often not achieve sufficient temporal and spatial resolution to compensate for the increased demands on size (up to 10x smaller) and heart rate (up to 10x faster) in these small animals. Improvements in small animal imaging – be it at the macro-, micro-anatomical, physiological or molecular level – have resulted in a progressively enhanced insight into the pathophysiology underlying dissecting AAA formation. In the current review article, we aim to provide an extensive overview of the 2D and 3D techniques that have been used to visualize different aspects of dissecting AAA formation in Angiotensin II-infused mice throughout the last 15 years. We discuss results obtained from ex vivo macroscopic observations, ex vivo histopathology and immunohistochemistry, in vivo high- frequency ultrasound, in vivo contrast-enhanced micro-CT and in vivo micro-MRI. We give an overview of how dissecting AAA in these mice has been detected, characterized and interpreted, discuss advantages and disadvantages of each technique and merge observations coming from researchers with different backgrounds, professional and technical expertise. We conclude with our recent advancements in the field using ex vivo PCXTM and PCXTM- guided histology (2D). We relate our findings to previous observations in the field, and demonstrate how the existing paradigm fits within the newly proposed one. Methods This narrative review is based on bibliographic data available on MEDLINE and PubMed up to December 2014. Used search terms were: “Angiotensin II, imaging, ultrasound, micro-CT, MRI and PCXTM” in combination with “abdominal aortic aneurysm, pathophysiology and mouse model”. We describe all imaging techniques that provide insight into Ang II-induced aneurysms, irrespective of the mice genotype (e.g. ApoE-/-, LdL-/-, or wild type). Aneurysms that were induced by combined Angiotensin II-infusion and intraperitoneal injection of anti-TGF-β or BANP were also included in the analysis. A total of 194 manuscripts was analyzed. Results Ex vivo macroscopic and microscopic anatomy In their seminal paper from 2000, Daugherty et al. were the first to report that a bulbous aortic abdominal shape occurred in 20% and 33% of ApoE-/- mice infused with 500 and 1000 ng/min/kg of Ang II, respectively (2). Already in 2001, Daugherty et al. classified Ang II-induced aneurysm morphology into 4 grades: Grade I was defined as a dilated lumen in the supra-renal region of the aorta with no thrombus, grade II as a remodeled tissue in the suprarenal region that frequently contains a thrombus, grade III as a pronounced bulbous form of type II that contains a thrombus, and grade IV as a form in which there are multiple aneurysms containing a thrombus, some overlapping, in the suprarenal area of the aorta (3) (Fig. 1). Fourteen years later, 134 of the 194 analyzed manuscripts have published pictures showing the gross morphology of the abdominal aortic aneurysm. As described above, aneurysm incidence is often based on macroscopic evaluation: either via a qualitative pathological evaluation of the morphology (4-18), or via a quantitative analysis of the external diameter that is obtained from caliper measurements on either the aneurysmal tissue itself or a digitized picture of it (19-25). Ex vivo detailed anatomy – histology Most of the observations that have contributed to our understanding of the pathology of murine dissecting AAA formation have been made by histological or immunohistochemical evaluation of the specimens. Since 2001, 112 studies published pictures of stained sections in dissecting AAAs, and 21 of these studies were designed specifically to describe dissecting AAA formation in Angiotensin II-infused mice. In the current review, we have chosen to focus on those images in which histology added insight into the existing paradigm of anatomy and pathophysiology in Ang II-induced dissecting AAAs. General dissecting AAA anatomy In their first publication, Daugherty et al. mentioned the existence of two regions of distinct characteristics within murine dissecting AAAs (2). The proximal dissecting AAA region had an intact cross-sectional lumen area and intact elastin layers in the media, but was associated with pronounced remodeling in the adventitial layer. The distal dissecting AAA region was characterized by a complete medial rupture that resulted in marked dilation of the lumen (2) (Fig. 2a). In 2003, Saraff et al. made a major contribution to the field as they observed that the early phase of Ang II infusion was characterized by a dissection and consequent vascular hematoma that occurred 4-10 days after the onset of infusion (26). They confirmed that a rupture occurred in the tunica media of the distal part of the dissecting AAA, and further reported that the bleeding was constrained by adventitial tissue in most mice, although in some animals transmural arterial rupture led to hemoabdomen and sudden death. Nevertheless, the authors concluded that after 28 days of Ang II infusion, aneurysmal tissue had many features of the human disease, including luminal dilation, extracellular matrix fragmentation, leukocyte accumulation, and thrombus (26). In 2011, Rateri et al. serially sectioned the full length of several dissecting AAAs, and stained every 10th slice (de facto staining at 100 µm intervals) with Movat’s pentachrome. They observed luminal dilatation and mural rupture of elastin fibers in the central part of the lesion, which was the only part they termed ‘AAA’ (27). Both proximally and distally of the dilated segment, a region with unchanged aortic lumen was observed adjacent to thickened adventitial regions that contained fibrous material and clotted blood, but was not considered part of the dissecting AAA (27) (Fig. 2b). The authors concluded that continuous infusion of Ang II into ApoE-/- mice led to gradual luminal expansion of the suprarenal aorta and that studies could be designed to test effects of interventions on aneurysm progression using this experimental model. Meanwhile, however, some authors questioned the term ‘aneurysm’ for a cardiovascular phenotype that does not show any consistent primary luminal dilatation. In 2007 Jiang et al. reported that Ang II infusion led to aneurysm-like expansive lesions that could not be classified as true aneurysms, as they were essentially encapsulated extramural thrombi (pseudoaneurysms)(28). They also reported a rupture of the tunica media in the distal part of the lesion, and postulated that the distal rupture in the tunica media may be associated with the proximal thrombus, confirming the original hypothesis by Saraff et al. (26). In 2012 Schriefl et al. performed serial histological sectioning of Ang II-induced dissecting AAAs (29). Similar to earlier reports, they observed that the central part of the dissecting AAA was characterized by completely ruptured elastic fibers and smooth muscle, while the distal and proximal ends of the dissecting AAA showed an intact true lumen adjacent to an intramural thrombus. Contrasting earlier reports, however, they also referred the existence of a third region, located just distally and proximally to the site of the dilated lumen (Fig. 2c). This region was characterized by a separate channel that ran parallel to the intact lumen, and later on joined the latter at the central, dilated part of the dissecting AAA (29). Interestingly, a similar parallel channel was visible on published images by Rateri et al. (27), but not classified as such (Fig. 2b). A similar structure can also be observed (often without interpretation) on published histology pictures by other authors (11, 30- 34) (Fig. 4b, 5). To discriminate murine AAAs from human AAAs (which demonstrate a true, luminal dilatation), Schriefl et al. used the term ‘dissecting AAA’ to describe the pathophysiology of this particular mouse model, which is also the preferred term in the current manuscript. Ruptured tunica media leading to luminal dilatation As the ruptured tunica media corresponds best to the luminal dilatation observed in human AAA, most reports in literature include histology images of this region. Disruption of the media has been associated with fragmentation of the elastic lamellae (2, 26, 33, 35-40), apoptosis of the smooth muscle cells (26, 29, 41-44) and infiltration of a variety of inflammatory cells (11, 19, 20, 26, 40, 44-53). The ruptured media occurs in the suprarenal part of the aorta (15, 17, 19, 26, 32, 37, 53-56), and is consistently found within the region of maximum dilatation (17, 19, 29, 37, 53, 57). In 2012, Gavish et al. were the first to report that transmural disruptions of the tunica media occur throughout the entire aorta in Ang II-infused mice, with a distinct predilection for branch orifices (36). They further demonstrated that at such sites of transmural disruption the extent of macrophage infiltration was increased and an attempted repair by newly deposited collagen could be observed (36). In follow-up research from 2014, Gavish et al. performed serial histology near the opening of celiac, mesenteric and both renal arteries(37). Transmedial disruptions occurred in 126 of 325 sections that included at least part of a branch orifice but, surprisingly, in none of the 479 sections without branch points. These disruptions ranged from small focal lesions to larger defects up to 1000 µm. There was no significant difference in branch-related transmedial disruptions between Ang II-infused mice that developed suprarenal dissecting AAAs and those that did not (37). Remodeled aortic wall adjacent to intact lumen A dilated wall adjacent to an intact tunica media has been reported by many authors (2, 18, 26, 27, 32, 33, 37, 58-63). In most publications, the dilated part was interpreted as an intramural thrombus (26-29, 32, 40, 46, 63, 64) or a remodeled adventitia (2, 31, 61, 65). Jiang et al. observed many infiltrating leukocytes within the thrombus after 4 weeks, indicating its organization (28). Cao et al. used Movat’s pentachrome staining to demonstrate remodeling events of collagen within the dilated aortic wall (61). Schriefl et al. observed deposition of glycosaminoglycans (GAGs) and fibrillar collagen in regions devoid of fibrin, which implies a “remodeling” process, typical or hematoma resorption and organization. The authors further speculated that the youngest thrombus existed closest to the center of the lesion where the merged lumen could have continued to provide flowing blood and hence fibrinogen and platelets, whereas the oldest region of the same thrombus was abluminal hence far from the flowing blood (29). While gross anatomy and detailed histopathology evaluations have provided substantial insight into the pathophysiology of dissecting AAA, a number of questions (such as the reason for the suprarenal location and large variation in aneurysm morphology) remained unaddressed.