Twenty Years of Studying AngII-induced Abdominal Aortic Pathologies in Mice – Continuing Questions and Challenges to Provide Insight into the Human Disease

Abstract

Angiotensin II (AngII) infusion in mice has been used to provide mechanistic insight into human abdominal aortic aneurysms (AAAs) for over two decades. This is a technically facile animal model that recapitulates multiple facets of the human disease. Although numerous publications have reported AAAs with AngII infusion in mice, there remain many fundamental unanswered questions such as uniformity of describing the pathological characteristics and which cell type is stimulated by AngII to promote AAAs. Extrapolation of the findings to provide insight into the human disease has been hindered by the preponderance of studies designed to determine the effects on initiation of AAAs, rather than a more clinically relevant scenario of determining efficacy on the established disease. The purpose of this review is to enhance understanding of AngII-induced abdominal aortic pathologies in mice, thereby providing greater insight into the human disease.

Graphical Abstract

Introduction

Abdominal aortic aneurysms (AAAs) continue to have a major impact on public health. While there have been advances in the surgical management of individuals afflicted with AAAs,¹ no medical interventions have been validated to attenuate expansion, decrease surgical interventions, or prevent rupture.^{2, 3} Consequently, there is a vital need for further research to provide mechanistic insight to facilitate determination of optimal therapeutic targets. The insurmountable barrier to acquiring human abdominal aortic tissues during the evolution of AAAs elevates the importance of acquiring meaningful data from animal models of the disease. It is now over 20 years since subcutaneous infusion of angiotensin II (AngII) was discovered serendipitously to promote AAAs in hypercholesterolemic mice.^{4, 5} Concurrently, there were publications of two other models of AAAs: one involved the intraluminal infusion of porcine pancreatic elastase,⁶ and the other one incubated a solution of calcium chloride on the adventitial surface.⁷ There have been several modifications to these models since the initial publications.⁸ Collectively, the availability of these mouse models contributed to the greatly augmented focus on research of AAAs over the last two decades. Since these initial publications, several reviews have comprehensively summarized many mechanistic insights of AAAs in these animal models.^9–12 This review is not intended to summarize the vast number of studies on AngII-induced AAAs that have been performed by many laboratories. Several comprehensive reviews have already summarized these data.^13–16 Instead, we will focus on several aspects that are salient and persistent questions with the hope of facilitating experimental design and interpretation.

Salient Features of Mice Infused with AngII to Promote AAAs

The experiments that led to the realization that AngII infusion promotes AAAs were performed to determine the effects of increased blood pressure on atherosclerosis, with AAA development being an unanticipated pathology. Hence, pressor and non-pressor infusion rates of AngII were administered to LDL receptor−/− mice fed a Western diet or Apoe−/− mice fed either normal or Western diet.^{4, 5, 17} Since these initial publications, we have learned features of experimental designs that influence the incidence and characteristics of aortic pathologies in mice infused with AngII.^18–20

Form of hypercholesterolemia.

The initial publications of AngII-induced AAAs used LDL receptor−/− mice fed a Western diet or Apoe−/− mice fed either a normal laboratory diet or a Western diet.^{4, 5, 21} Based on publications that have reported AngII-induced AAAs in both LDL receptor−/− and Apoe−/−mice, it appears to be more severe disease in the latter, ^{22, 23} although no direct comparisons between these two strains have been reported to confirm this speculation. While it has been consistent that hypercholesterolemia augments the incidence of AngII-induced AAA formation, the AAAs that form under normo- and hypercholesterolemic conditions appear to have the same characteristics. In normocholesterolemic mice, the incidence of AngII-induced AAA is low; generally in the range of less than 20%,^{5, 23–25} although higher incidences have been reported.²⁶ This contrasts with incidences in hypercholesterolemic mice that are generally greater than 80%. The augmentation of hypercholesterolemia has been demonstrated directly in C57BL/6J mice that were rendered hypercholesterolemic acutely through infection with an AAV expressing a gain-of-function mutant of proprotein convertase subtilisin/kexin type 9 (PCSK9).^27–29 The use of AAV expressing mutants of PCSK9 has been shown to augment AngII-induced AAA formation.^{30, 31} The extent of hypercholesterolemia needed to augment AAA formation seems relatively low. For example, Apoe−/− mice fed a Western diet that increased plasma cholesterol concentrations up to 1,000 – 1,500 mg/dl had no difference in AAA incidence compared to mice fed a normal laboratory diet with plasma total cholesterol ~ 400 mg/dl.²³ In LDL receptor −/− mice, withdrawing Western diet reduces plasma total cholesterol to < 400 mg/dl and halts expansion of AAAs during prolonged AngII infusion.³² Therefore, hypercholesterolemia is needed to maintain the propagation of aneurysmal expansion. Despite the uniform demonstration of the augmentation of hypercholesterolemia on AngII-induced AAAs, the mechanism of this effect is unclear.

Age.

Some of the earlier publications infused AngII into mice that were 6 to 11 months of age. However, these ages were the consequence of mouse availability rather than any adjudicated aspect of experimental design. Although age is a major factor in AAA development in humans, there is only one publication in which the effect of age on AngII-induced AAAs has been formally addressed. This study compared 2–3 versus 18–20-month-old C57BL/6J mice and showed a greater AAA incidence in the older mice.³³ While there needs to be confirmation of the age-related effect, the vast majority of publications describe studies in which AngII infusion was initiated soon after sexual maturity has been attained (i.e., 2 months of age).

Sex.

While the initial studies on AngII-induced AAAs were performed in females,⁵ there has been a consistent demonstration of sexual dimorphism with a much greater incidence of disease in males.^34–37 Gonadectomy studies have demonstrated a clear effect of removal of testes in reducing AAAs, while removal of ovaries has minimal effect.³⁶ The sexual dimorphism is attributed to both hormonal and chromosomal influences, with more recent studies suggesting protection of females with two X chromosomes from AngII-induced AAAs.^{20, 37, 38} While the mechanisms of the sexual dimorphism are unclear, the clear divergence of responses exemplifies the need to report data in a sex-specific manner.³⁹

Blood Pressure.

A pressor effect was demonstrated at an infusion rate of AngII that is used commonly in AAA studies (most stated as 1,000 ng/kg/min, but some stated as 1.44 mg/kg/d). However, increasing blood pressure by norepinephrine infusion to an equivalent level attained during AngII infusion did not promote AAA development.¹⁷ Conversely, decreasing systolic blood pressure by hydralazine administration did not reduce AngII-induced AAAs.¹⁷ Furthermore, other interventions have dissociated AngII-induced AAA formation from changes in blood pressure. For example, the smooth muscle cell-specific deletion of RelA decreased AngII induced AAA without any change in blood pressure.⁴⁰ Similarly, castration of male mice markedly reduced AngII-induced AAAs while having no effect on blood pressure.³⁶ A summary of selected publications that reported dissociation or association between systolic blood pressure and AngII-induced AAA formation is presented in Table I. Overall, the literature does not support that the AngII-induced increase in blood pressure is a major contributor to AAA formation in mice.

Table I.

Selected examples of interventions in which blood pressure changes were either dissociated or associated with changes in AngII-induced AAA formation.

Intervention	Effect		Reference
	Blood Pressure	AAA
Dissociated
Norepinephrine	↑	No AAA	17
Prostaglandin E receptor 4 transgenic	↔	↑	105
Acute lung injury	↔	↑	106
Hydralazine	↓	↔	107
IL27 receptor -/-	↔	↓	51
IL-18 -/-	↔	↓	108
Mas receptor -/-	↔	↓	84
NF-κB/RelA	↔	↓	40
P47phox -/-	↔	↓	109
Fasudil	↔	↓	110
β-Arrestin-2	↔	↓	111
Diltiazem	↔	↓	112
Iron restriction	↔	↓	113
Nlrp3 or Casp1 -/-	↔	↓	114
BAF60a -/- in SMCs	↔	↓	115
Vitamin D	↔	↓	116
Paraoxonase genes transgenic	↔	↓	117
microRNA-33 -/-	↔	↓	118
Cilostazol	↔	↓	119
SGLT-2	↔	↓	52
Bmal1 -/-	↔	↓	120
Low dose nifedipine (5 mg/kg/day)	↔	↓	121
Associated
Endothelial Gch1 -/-	↑	↑	122
Nhe1	↓	↓	49
High dose nifedipine (20 mg/kg/day)	↓	↓	121

Note: ↔: no effect, ↓: decrease; ↑: increase

Infusion Rate.

In initial studies, LDL receptor−/− mice⁴ or Apoe−/− mice⁵ were infused with AngII at 500 ng/kg/min or 1,000 ng/kg/min. Most later studies infused AngII at 1,000 ng/kg/min to male mice, with a few studies reporting infusion rates of 2,500 ng/kg/min. One study using this higher AngII infusion rate reported a higher incidence of AAAs compared to 1,000 ng/kg/min in normolipidemic mice.⁴¹ However, these two rates have not been compared in hypercholesterolemic mice side-by-side. Since the incidence of AAAs is approximately 60 – 100% in hypercholesterolemic mice, it is possible that an infusion rate above 1,000 ng/kg/min will not further increase incidence of AAAs.

Characteristics of the Pathology of AngII-induced AAAs

AngII infusion promotes rapid region-specific changes of aortic pathology.^{19, 42} The challenges to characterize the pathological characteristics of the tissue include variable manifestations in disease presentation between individual mice, the striking heterogeneity of the aneurysmal tissue, and the changing characteristics during continued AngII infusion.

One characteristic of AngII-induced AAA tissues is considerable spatial and temporal variations of pathologies along the length of the aneurysmal tissues (Figure 1).^{42, 43} AAA formation during AngII infusion occurs in the supra-renal aorta, which has been a consistent feature of several genetic and chemically induced models.^{44, 45} Currently, there is no compelling evidence that explains this localization. Based on acquisition of abdominal aortas after selected intervals of AngII infusion, macrophages were detected in the media within 2–3 days of AngII infusion. Given the medial location the sources of macrophages could be either blood-borne or resident in the adventitia. The most apparent changes in the first few days is a rapid lumen expansion in which the dilated region has overt thrombus.⁴² These large adventitial thrombi may be derived by the microruptures that have been demonstrated at the ostium of the celiac and mesenteric arteries.⁴⁶ Histology and immunostaining of tissue sections in this thrombus region display distinct heterogeneity that requires sectioning throughout the affected region to acquire an appreciation of the regional changes. Many of the tissues acquired within days of initiating infusion have large thrombi located outside the external elastic lamina and therefore being exclusively present in the adventitia. While an adventitial thrombus encasing the intact media is a predominant pathology, a large luminal expansion in some sections is present consistently. A major characteristic of the tissue sections at the luminal expansion is a transmural break of the elastic fibers in the media, with the thrombus being constrained by adventitial collagens. This pathology is consistent with a rapid focal loss of medial integrity resulting in a localized aneurysm and an adjacent area of thrombosis caused by adventitial dissection. At this early interval, most publications do not provide evidence of pronounced changes in the elastic fibers beyond this transmural break. The rapidity of lumen expansion was subsequently verified with the advent of high frequency ultrasound that enables sequential measurements of aortic diameters in living mice. Use of this modality demonstrated that AngII infusion increases lumen diameter from ~ 0.9 mm to ~ 1.5 mm within a week.^{47 48} The rapidity of the onset of pathology was also demonstrated by the occurrence of death due to rupture of the abdominal aorta that frequently occurs within the first 7 days, and then becomes infrequent after this interval.^49–52 Sequential ultrasound imaging has demonstrated that large increases in diameter can be detected in mice during this interval of AngII infusion, prior to rupture. Some mice subsequently succumb to loss of aortic integrity several days after the detection of an expanded lumen.⁴⁷ This may be interpreted that the transmural medial rupture does not immediately compromise the aortic integrity. Instead, there is a loss of aortic integrity subsequent to the acute luminal expansion that leads to rupture-induced death. A characteristic of the thrombus that forms in mice succumbing to abdominal aortic rupture is the proclivity to be localized in the left retroperitoneal region.⁵³ The nature of the AAAs is also commonly depicted as being predominantly left sided. The speculation for AngII-induced AAAs having a left sided predominance has been suggested to be anatomical and hemodynamic.^{53, 54}

Figure 1. Diagrammatic representations of the temporal and spatial changes during evolution of AngII-induced AAAs.

A. Changes in diameter and appearance of the suprarenal aorta during 84 days of AngII infusion. Within 7 days of initiating AngII infusion, there is a rapid expansion of the luminal diameter of the suprarenal aorta (red). Subsequently, there is gradual progressive expansion as reported up to 84 days of AngII infusion (green). B. Spatial changes during 28 days of AngII infusion. There is considerable heterogeneity of pathology along the length of the suprarenal aorta during AngII infusion. An early event is focal medial rupture that leads to exit of blood to provoke adventitial dissection, resulting in two major pathological manifestations: 1. A region of lumen expansion. 2. An adjacent area in which the lumen diameter is normal but there is pronounced adventitial thickening. The thrombotic material is gradually remodeled to extracellular matrix. These temporal and spatial changes are represented in the accompanying movie (Online Video I).

An early cellular characteristic of the aorta following AngII infusion is accumulation of macrophages in the adventitia abutting the external elastic lamina.⁵⁵ There have been reports that medial accumulation of macrophages occurs as an initial cellular event, although this needs to be substantiated.^{19, 42} Following the initial rapid initiation phase, there is an accumulation of T and B lymphocytes and neutrophils in AngII-induced AAAs.^56–60 Following the transmural medial rupture, there is profound accumulation of macrophages that coat the adventitial thrombus.⁴² The source of these macrophages has not been defined, although the large number of cells is consistent with being derived from the blood. With continued AngII infusion, there is a gradual resolution of the thrombus that is replaced by an amorphous material showing high abundance of extracellular matrix protein. During this phase, two major areas are distinguishable. One has an intact media of normal dimensions with a thickened adventitia.⁵ These regions are proximal and distal to the aneurysmal area in which the adventitial dissection initially leads to thrombus accumulation. The other major area has a grossly expanded lumen. In the earlier phases, elastic layers are only present in the region of the original lumen before transmural medial rupture. With protracted AngII infusion to male mice, there is increased accumulation of macrophages that is enhanced in the regions with elastin fragmentation.⁴³ With further infusion, the elastic fibers form over the entirety of the expanded lumen. Atherosclerotic lesions can also be detected in AAA tissue sections from hypercholesterolemic mice infused long-term with AngII. With prolonged infusion, the aortic wall becomes thinner and aortic rupture may occur in this late phase.^{43, 61}

The complexity of the pathology formed during AngII infusion creates difficulties in objectively defining tissue characteristics. To simplify the analysis and reporting, there are two major distinctive pathologies as noted in the original publication.⁵ One region has pronounced expansion of lumen diameter associated with a transmural medial break. The other major tissue characteristic is an adjacent region in which medial diameter and structure are grossly normal with substantial adventitial thickening. The difference between these two distinct pathologies becomes less apparent with continued AngII infusion beyond 4 weeks.⁴³ At present the maximum duration of AngII infusion that has been reported is 3 months.⁴³ Overall, meaningful interpretation of tissue sections of AngII-induce aneurysm needs cognizance that there are profound spatial and temporal differences and that meaningful conclusions can only be drawn from studies that account for these differences.

Progressive Expansion of AAAs during Continued AngII Infusion

The dimensions of AngII-induced AAAs can be acquired on ex vivo tissues in which the maximal external diameters are measured. The availability of high frequency ultrasound with resolution in the range of 30 μm has permitted sequential measurements of luminal diameter during AngII infusion in vivo. Unlike measurement of human AAAs, the measurement of AngII-induced AAAs is most frequently represented as the maximal lumen diameter and not the external dimensions. The sequentially acquired aortic dimensions in vivo confirmed the data from destructive analysis that there is a rapid increase in lumen diameter within 7 days of AngII infusion being initiated.⁴⁷ Following this initial rapid expansion, there is a relatively modest increase for the remainder of the 28 days. Twenty-eight days is the most common duration for these studies since the vast majority of the studies use the Alzet models 1004 or 2004 that have a 28-day infusion period. However, it is a relatively trivial and well tolerated surgical procedure to replace the Alzet pump to perpetuate AngII infusion. Despite the ease of prolonging AngII infusion, there have been relatively few studies that have infused AngII for more than 28 days. Those studies in which prolonged infusion has been continued for a further 2 months have demonstrated that in LDL receptor−/− mice fed a Western diet or Apoe−/− mice continued infusion of AngII promotes a gradual increase in lumen dimension expanded to ~ 3 mm after 3 months of AngII infusion.^{32, 43, 61–63} Continuous expansion only occurs when there is simultaneous continuation of both AngII infusion and feeding a Western diet in LDL receptor−/− mice.^{32, 62} Marked expansion has been noted in aortic diameter of Apoe−/− mice that were infused for 28 days followed by no further infusion for another 8 days, demonstrating the irreversibility of lumen dilation.⁶⁴ The aortic diameter was increased by infusing AngII for 28 days in C57BL/6J mice fed a high fat diet. Subsequent withdrawal of the high fat diet did not decrease aortic diameters when AngII infusion was continued for a further 56 days.²⁴ These findings are consistent with hypercholesterolemia, and/or obesity, being critical for progression of AngII-induced AAAs. As relates to sex differences, castration of male mice prevented the progression of established AngII-induced AAAs, while administration of exogenous 17β-estradiol to castrated females blunted progression of AAAs.⁶³

The determination of the experimental conditions leading to continued expansion will be important for studies that look at the effects of interventions on progression of established AAAs. Meaningful conclusions of these studies need a definable parameter that changes in a robust manner over time. There is a consistent literature demonstrating that AngII infusion into LDL receptor−/− mice fed a Western diet have discernable increases in aortic diameter and aneurysmal tissue characteristics during prolonged infusion.^{32, 62} The research arena would be assisted by a greater focus on conditions of aneurysmal propagation that is relevant to the human disease.

Different Response of Therapeutics in Inhibiting Initiation versus Propagation of Established Disease

There may be distinct mechanisms that determine the different phases on the aortic tissues including the initiation, propagation, and rupture. A large number of studies with AngII-induced AAAs have studied a genetic or pharmacological intervention that was present at the time of initiating the AngII infusion. However, in the clinical setting, therapies would only be provided to those individuals that have established diseases. Hence, it is important that more studies shift the emphasis to efficacy of interventions in established AAAs, thereby providing relevance to humans. For example, doxycycline was proved to be effective when administered prior to the initiation of AAA formation during AngII infusion, and in the elastase and calcium chloride models of the disease,^{6, 65, 66} although negative data were also reported.⁶⁷ Despite the relative consistency of these responses, doxycycline failed to modify AAA expansion in humans.^{68, 69} Of note, doxycycline also failed to modify AngII-induced expansion when the drug was administered to mice with established AAAs⁶² at a dose that led to the plasma concentrations of the drug considered to effectively inhibit MMP in the N-TA³CT trial.

While it is desirable to perform interventions to determine their efficacy on the established disease of AngII-induced AAAs, there are several impediments to such studies. One issue is that many of the studies have been performed using constitutive genetical deletions, and there are significant investments needed to develop mice in which these studies could be repeated in mice with conditionally activated genes after induction of AAAs. For pharmacological interventions, there are also significant impediments. An optimal execution of these studies requires infusion of AngII to form AAAs, followed by ultrasonic measurements of aortic diameter to enable stratification of mice, and continued infusion to permit a sufficiently robust expansion in the control group that meaningful statistics can be determined for the pharmacological intervention.⁷⁰ Given that not all mice infused with AngII develop AAAs, these studies require significant numbers of mice/group and careful stratification of mice with established AAAs (which differ in size from mouse to mouse) to control versus intervention groups.

One major decision in experimental design that could influence the outcome of an intervention is the timing following AngII infusion. As noted above, AngII promotes a rapid expansion within 7 days with the predominant pathology in the region being thrombus. The abdominal aorta then progresses to profound remodeling that results in dramatically different characteristics of aortic wall at different intervals after infusion. Therefore, the pathological status of the aneurysmal tissue will need to be a major consideration in determining the time of onset of an intervention.

Receptors Stimulated by AngII to Form AAAs

In recent years, there has been identification of an increasing number of receptors that can be activated by angiotensin peptides.⁷¹ However, for AngII, the major receptors are AT1 and AT2. In rodents, there has been a chromosomal duplication process leading to this species expressing two isoforms of AT1 receptors, termed AT1a and AT1b. The aorta expresses both isoforms.^{72, 73} These receptors are of 94% identity on amino acid sequences which can create challenges to discriminate for determination of their abundance at the protein level. Unfortunately, it is widely assumed that angiotensin receptor antibodies lack the ability to detect the endogenous protein,^{74, 75} and the close amino acid similarity of the isoforms creates an additional hurdle to detect the presence of these receptor proteins.

AT1 receptor blockers can effectively antagonize both AT1a and AT1b receptor isoforms. The inaugural member of this class of drugs, losartan, ablates the effects of AngII on AAA formation, consistent with the pathology being due to one of the AT1 receptor isoforms.²¹ This protection has also been demonstrated for telmisartan, irbesartan and valsartan.^{67, 76} Genetically manipulated mice are available to determine the angiotensin receptor subtype responsible for the pathology. Deletion of AT1b receptors did not affect the development of AngII-induced AAAs,⁷⁷ while deletion of AT1a receptor ablates AAA formation in LDL receptor−/−mice.⁷⁸ mRNA for both receptors are present in the aorta, although detection of the protein by antibodies has been hindered by the lack of suitable antibodies.^{75, 77} The different responses to deletion of AT1a versus AT1b receptors may be attributable to the small number of variances in the cytoplasmic domains of the two receptors that influence signaling pathways.⁷⁹ Interestingly, these pathological effects on aortic pathology are not mimicked by the contractile physiologic response to AngII by this tissue. Incubation of aortic rings from selected aortic regions with AngII only results in contraction of the infra-renal aorta that does not display profound pathology after AngII infusion. Deletion of AT1a receptors has no effect on aortic contraction in this region, while deletion of AT1b receptors ablates this response.^{72, 73, 77}

AT2 receptors are also activated by AngII in a mode that generally opposes the actions of AT1 receptor stimulation.⁸⁰ These receptors are highly expressed in various tissues at the fetal stage, but are weakly expressed in tissues of the adult.⁸¹ Despite the modest AT2 receptor expression in the aorta, administration of the AT2 receptor antagonist, PD123319, markedly increased the severity of the AAAs in AngII-infused Apoe−/− female mice.²¹ However, subsequent studies demonstrated that AT2 receptor genetic deficiency had no effect on AngII-induced AAAs. Indeed, PD123319 also augmented AngII-AAAs in AT2 receptor −/− mice, demonstrating that the effect of PD123319 was independent of AT2 receptor antagonism.⁸² PD123319 is now known to antagonize receptors other than AT2 receptors,⁸³ although its mechanism of augmenting AAAs remains to be clarified.

The only other angiotensin receptor with a demonstrated role in AngII-induced AAAs is the Mas receptor that is stimulated by Ang1–7. Deficiency of this receptor augmented AngII-induced AAAs that is consistent with Ang1–7 being a protective angiotensin peptide during the disease evolution.⁸⁴

Overall, there has been a uniform demonstration that AngII exerts its effects on the formation of AAA through activation of AT1a receptors. This clarity will facilitate the further definition of the role of the complex signaling pathways that are stimulated by AngII activation of AT1a receptors and the functional consequences.⁸⁵

Cell Types Stimulated by AngII to Promote Abdominal Aortic Disease

Given the consistent reports of the role of AT1a receptors in AngII-induced AAA development, insight would be derived from determination of the cell type being stimulated by AngII. The availability of global AT1aR−/− mice enabled determination of the role of this receptor in leukocytes using the approach of bone marrow transplantation. The studies were performed by irradiating both AT1a receptor +/+ and −/− mice and repopulated with either AT1a receptor +/+ or −/− bone marrow-derived cells. These studies failed to demonstrate a role of the AT1a receptor on leukocytes in aneurysm development.⁷⁸ With development of AT1a receptor floxed mice, this has permitted the interrogation of cell types using cell-specific promoters.⁸⁶ Given that the major cell type in aneurysmal tissue is smooth muscle cells, deletion of the receptor in these cells was expected to alter aortic pathology. However, the first study using these mice demonstrated that deletion of AT1a receptors using SM22 Cre failed to affect AngII-induced aortic pathology.⁸⁷ In AT1a receptor floxed mice that were developed in C57BL/6J derived stem cells and bred to SM22 Cre expressing mice, there was also a lack of any discernable effects on AngII-induced AAAs in LDL receptor−/− mice.⁸⁸ These floxed mice were also bred to mice expressing Cre under the control of the Tie2 promoter to delete AT1a receptors in endothelial cells. This deletion also has no effect on development of AngII-induced AAAs.⁸⁸ Therefore, based on the current literature, there is no evidence thus far that AngII promotes development of AAA through direct interactions with AT1aR on leukocytes, smooth muscle cells, or endothelium.

Targeting of two other cell types have demonstrated a possible role of adventitial cells on AngII-induced AAAs. One cell type is white adipocytes which surround the region of AngII AAA formation.²⁵ A direct role of AT1a receptor in adipose tissue has been implicated using mice in which adipose tissue from either AT1a receptor +/+ or −/− mice was transplanted into Apoe −/−mice expressing AT1a receptors.⁸⁹ Transplantation of adipose tissue with AT1a receptor deficiency led to a significant reduction in the diameter of the abdominal aorta following AngII infusion. Adipose transplantation was also performed in Apoe−/− mice that were AT1a receptor deleted. Unlike the effect of the donor adipose tissue in the AT1a receptor wild type mice, transplantation of AT1a receptor wild type adipose tissue had no effect on AAA formation in recipients with a AT1a receptor −/− genotype.⁸⁹ Therefore, the role of direct stimulation of AngII on adipocytes in the generation of AAAs needs confirmation and clarification. The other cell type that has been implicated in AngII-induced AAAs is fibroblasts. A role for this cell type was implied by the demonstration that AngII augments monocyte chemoattractant protein-1 and interleukin-6 from adventitial fibroblasts.⁹⁰ Consistent with this implication, deletion of an AngII responsive signaling pathway in fibroblasts attenuated development of AngII-induced AAAs.⁴⁰ While there is a report that fibroblast-specific deletion of AT1a receptors attenuated aortic pathologies,⁹¹ there has not been a direct determination of a role in AngII-induced AAAs.

There has been a consistent literature that AT1a receptor genetic deletion or pharmacological blockade abrogates AngII-induced AAAs. However, there is a lack of clarity on which cell type is being stimulated by AngII. There is also possibility that cluster of cells needs to be stimulated by AngII to enable the development of AngII-induced AAAs. Definition of the cell type(s) involved in the development of AAAs would provide a major step forward to refining the mechanism of disease development.

Insights from AngII-induced Aortic Disease in Mice to the Human Disease

A consistent feature of AngII-induced AAAs is the supra-renal location. Of note, aneurysms are formed in the same location of several other mouse models, including LDL receptor−/−,⁹² Apoe−/−,⁹³ compound Apoe−/− × endothelial nitric oxide synthase−/−,⁴⁴ and mice administered a mineralocorticoid receptor agonist and salt.⁴⁵ This contrasts the most common location of the human disease in the infrarenal aorta. There are several speculations for the location of the human disease including regional differences in the composition of the extracellular matrix and adventitial layer, the lack of medial vasa vasorum in this region, hemodynamic stresses, and the distinct embryonic origins of smooth muscle. It is unknown whether the regional contrast in AAA location between humans and mice is attributable to any of these possibilities. One of the most obvious differences is humans being biped while mice are quadruped which is likely to impart regional hemodynamic and structural differences between the species. It should also be noted that infusion of AngII causes aortic pathology in other regions, including the ascending and distal thoracic aorta.^{37, 94–97}

The interest in determining a role of the renin angiotensin system in the experimental setting has also spawned interest in clinical research of AAAs. One indication of the interest in the renin angiotensin system has been the trials on inhibitors. This has included a trial (Aortic Aneurysmal Regression of Dilation: Value of ACE-inhibition in RiSK – AARDVARK) on the efficacy of the ACE inhibitor, perindopril on AAA expansion over 2 years compared to a control group provided with placebo and a third group given the calcium antagonist, amlodipine as a control for blood pressure lowering. As with the recently completed N-TA³CT trial,⁶⁹ the rates of AAA expansion were smaller than anticipated. While the trial failed to demonstrate any effect of ACE inhibitor, this should be evaluated in the context of the small rate of expansion in the control group (1.68 mm) over the interval of the trial (2 years) in a modest group size of patients (73 patients received perindopril). There is a recent study to determine the efficacy of telmisartan. This study did not show an effect of the administration of this angiotensin receptor blocker to inhibit AAA expansion. However, similar to the AARDVARK trial, it was also too underpowered to draw definitive conclusions.⁹⁸

A practical issue on determining a role in human is the commonality of inhibitors of the renin angiotensin system that are prescribed for hypertension. However, currently available inhibitors of the renin angiotensin system use low doses of drugs with short half-lives that may not provide sustained chronic inhibition at a level easily achieved in experimental studies showing benefits of blockade. A possible alternative is drugs in development to chronically inhibit synthesis of the unique angiotensin precursor, angiotensinogen^{99, 100}, or drugs targeted at other components of the renin-angiotensin system such as Mas receptors⁸⁴ and ACE2.¹⁰¹

What studies would provide further insight?

One of the recurrent questions of the AngII-induced AAA mouse model has been why the pathological changes are localized to the suprarenal region as compared to human infrarenal AAAs. There have not been any defined morphological medial characteristics in tissue sections of the supra renal area that provide clues to the susceptibility to aortopathy. However, the characteristics of the adventitial area have not been well defined, although it is known that adipocyte characteristics differ in mice between the thoracic and abdominal region.²⁵ The adventitia adjacent to human aorta has also not been well defined, either under normal or disease conditions.

Although there are no overt morphological characteristics of the suprarenal aorta that provide insight into its susceptibility for aortopathy, there may be more subtle differences that would not easily be discriminated by pathological techniques. As described above, infusion of AngII promotes aortic pathology in other regions of the aorta, with some recent studies identifying differences in genes expressed within different aortic regions in a manner consistent with pathology formation.^{37, 38} Currently, there have been no in-depth interrogations of the different regions using proteomic approaches which may provide insight into the disease location. There is also a lack of this type of information in the human disease.

There is still a surprising lack of information on how AngII promotes aortopathy in mice. For example, basic questions have not been addressed including how subcutaneously infused AngII reaches a site to promote the disease, particularly when the half-life of the peptide in plasma is only a few seconds.¹⁰² The identification of this site has currently been refractory to the approaches used that have primarily been bone marrow transplantation and cell specific deletions of the AT1a receptor which is responsible for the disease progression. Definition of this site would assist in the determination of which cell type derived from humans should be the focus of studies on AngII activation.

Perspectives

There is a voluminous literature demonstrating effects on AngII infusion, inhibitors of AngII production, and AngII receptor activation on several forms of aortic pathology. Although the focus of this review was on the changes in the abdominal aortic region, there have been recent data on the role of AngII in thoracic aortic disease.^{95, 103} The effects of AngII on this aortic region has rapidly progressed from animal studies to clinical trials.^{95, 103, 104} These effects on aortic pathology are not a property that is shared with other pressor agents.⁹⁴ The unusual consistency by which manipulations of the renin angiotensin system influences aortic pathology indicates the need for further research to enhance our understanding to determine their contributions to the development of AAAs.

Highlights.

1. Chronic subcutaneous infusion of AngII is a frequently used mode of developing abdominal aortic aneurysms (AAAs) in mice.

2. Pathological characteristics of AngII-induced AAAs are complex and heterogenous that change in spatial and temporal modes.

3. Application of the data from this model to the human disease may be enhanced by an increased focus of studies that determine effects of interventions on established AAAs.

Nonstandard Abbreviations and Acronyms

AngII

Angiotensin II

AAA

Abdominal Aortic Aneurysm

Apoe

Apolipoprotein E

LDL

Low density lipoprotein

PCSK9

Proprotein convertase subtilisin/kexin type 9