Apolipoprotein E knockout rabbit model of intracranial atherosclerotic disease

Abstract Intracranial atherosclerotic disease (ICAD) is the most common cause of ischemic stroke. Poor understanding of the disease due to limited human data leads to imprecise treatment. Apolipoprotein E knockout (ApoE‐KO) rabbits were compared to an existing model, the Watanabe heritable hyperlipidemic (WHHL) rabbit, and wild‐type New Zealand white (NZW) rabbit controls. Intracranial artery samples were assessed on histopathology for the presence of ICAD. Logistic and ordinal regression analyses were performed to assess for disease presence and severity, respectively. Eighteen rabbits and 54 artery segments were analyzed. Univariate logistic analysis confirmed the presence of ICAD in model rabbits (P < .001), while no difference was found between WHHL and ApoE‐KO rabbits (P = .178). In multivariate analysis, only classification as a model vs wild‐type animal (P < .001) was associated with the presence of ICAD. Univariate ordinal regression analysis demonstrated an association between ICAD severity and model animals (P = .001), with no difference was noted between WHHL and ApoE‐KO rabbits (P = .528). In multivariate ordinal regression analysis, only classification as a model retained significance (P < .001). ICAD can be reliably produced in ApoE‐KO rabbits, developing the disease comparably to the older WHHL model. Further analysis is warranted to optimize accelerated development of ICAD in ApoE‐KO rabbits to more efficiently study this disease.


| INTRODUC TI ON
Intracranial atherosclerotic disease (ICAD) is the most common cause of stroke worldwide. [1][2][3] Pathophysiological mechanisms of ICAD have not been clarified in a way that is specific to intracranial disease, so current clinical management strategies are suboptimal.
Better understanding of ICAD is needed through basic science and translational research to develop more effective treatment. A major hindrance in the study of ICAD pathophysiology is the lack of tissue for pathological evaluation, but this cannot be reliably obtained in humans. Biopsy is prohibitively morbid, so tissue is typically only obtained at autopsy.
Given limitations in human investigation of ICAD, current management is based on extrapolation from extracranial disease. Such extrapolation may be flawed since intracranial and extracranial vessels arise from different germ cell layers, ectoderm, and mesoderm, respectively, so it should not be assumed that these vessels and the diseases they may harbor are the same. [4][5][6] Several differences are known between intracranial and extracranial arteries, such as the presence or absence of vasa vasorum, size of adventitia, response to systemic processes like serum cholesterol or hypertensions, and their differential involvement in genetic disorders like familial hypercholesterolemia. 4,5,[7][8][9][10][11][12] Further evaluation of ICAD is needed, including comparisons to extracranial disease that test the appropriateness of the current extrapolation from extracranial to intracranial vessels. Given the paucity of human data, an animal model is critically needed.
Animal models for the study of atherosclerosis have been reported in species as small as mice and as large as nonhuman primates, with most work occurring in rodent and rabbit models. 13,14 Apolipoprotein E knockout (ApoE-KO) mice have been shown to develop intracranial atherosclerosis; this approach is based on understanding of the plasma lipoprotein's role in cholesterol degradation and association with premature atherosclerosis in humans when deranged. 15,16 While useful for certain investigations, large animal models are necessary for better assessment with noninvasive imaging and surgical procedures that translate to human care. 14,17 As in humans, most rabbit atherosclerosis research involves extracranial vessels. The most widely studied rabbit model of atherosclerosis is the Watanabe heritable hyperlipidemic (WHHL) rabbit. 14,18-20 WHHL rabbits develop ICAD, although some evidence exists that it typically only occurs after the induction of hypertension. 21,22 To date, most WHHL investigation has focused on extracranial disease, and the mutation underlying this strain is not associated with intracranial disease in humans. 14,[19][20][21][22] Additionally, in recent years, steady supply of these animals has not been maintained, so a better model of ICAD is needed. 23 Apolipoprotein E knockout (ApoE-KO) rabbits have been used for the evaluation of extracranial atherosclerosis, but they have not been evaluated for the presence of ICAD. 24 This study describes the comparisons of intracranial arteries in WHHL and ApoE-KO rabbits with comparison to NZW controls.

| ME THODS
All studies were conducted in compliance with a protocol approved by the institutional animal care and use committee. Mature WHHL and ApoE-KO rabbits fed a regular or custom atherogenic diet (0.3% cholesterol, 3% soy bean oil, Envigo Teklad Diets), as well as younger NZW rabbits, were also analyzed. 23 WHHL rabbits were sourced from the Shiomi Laboratory at Kobe University. ApoE-KO rabbits were provided by Horizon Discovery Group, which was subsequently acquired by Envigo, Indianapolis, IN, who also supplied NZW rabbits.
For euthanasia under general endotracheal anesthesia induced and maintaining with isoflurane, a midline neck incision was made over the trachea. The right carotid artery was isolated using blunt dissection; gentle tension was applied with vessel loops. A 22-gauge intravenous catheter was inserted into the artery. A 5-French micropuncture sheath was then placed in the artery over a 0.018 in microwire using Seldinger technique. This sheath was secured within the vessel with a 2-0 silk suture. After removing the inner dilator and wire, the sheath was connected to the perfusion pump to deliver 2% paraformaldehyde and 5% glutaraldehyde solution. After initiating perfusion, a jugular vein was transected to limit blood in the vessels and tissues after exsanguination. Following adequate perfusion, the animal was decapitated to harvest the brain and its intact arteries, which were placed in formalin.
After at least 2 weeks in formalin, tissue was sliced to maximize cross-sectional orientation of basilar and internal carotid arteries.
Hematoxylin and eosin staining was performed and slides were created and studied with light microscopy. Internal carotid and basilar artery segments were each rated as having no, mild, moderate, or advanced ICAD. Severity was determined by histological findings including smooth muscle hypertrophy, wall thickening, infiltration of inflammatory cells, neointimal formation, presence of lipids within neointima, and remodeling of the vessel lumen.
Descriptive statistics were performed, assessing disease burden on a by-segment basis. Further analysis was performed to compare animal features and disease burden. When assessing for the presence of disease, binary logistic regression analysis was performed.
When assessing for severity of disease, ordinal regression analysis was performed. Multivariate models were constructed, including variables with P < .010 in univariate analysis, to assess for confounding factors. Statistics were performed using R (R Foundation for Statistical Computing, Vienna, Austria).

| RE SULTS AND D ISCUSS I ON
Eighteen rabbits (5 WHHL, 5 ApoE-KO, and 8 NZW) and 54 artery segments underwent evaluation. Table 1   Most of the existing work on intracranial disease has occurred in WHHL rabbits. These animals lack low-density lipoprotein receptors, which in turn leads to the development of atherosclerosis. 14,18,19,27,28 Research on WHHL rabbits led to the discovery of the underlying mutation causing familial hypercholesterolemia. 19 ICAD typically does not occur in patients with familial hypercholesterolemia, and the overwhelming majority of ICAD patients do not suffer from this rare disease, so WHHL rabbits may not be an appropriate ICAD model. [29][30][31] Additionally, WHHL rabbits have poor general health.
ICAD is reported to occur most reliably only after hypertension induction; this may further complicate the health of these fragile animals and prevent reliable long-term analysis. 21,22 Other methods of atherosclerosis induction exist for rabbits. 14 ApoE is involved in lipid transport and has been targeted for disease models in multiple species; knockout rabbits can now be selectively bred. 24

ACK N OWLED G M ENTS
This study was funded in part by the American Heart Association

Transformational Grant 19TPA34910194 and The Joe Niekro
Research Grant from the Joe Niekro Foundation and Society of NeuroInterventional Surgery Foundation.

CO N FLI C T O F I NTE R E S T
None.

AUTH O R CO NTR I B UTI O N S
All the authors participated in animal care, including perfusion fixa-  Table 1). Weeks on the x-axis start on at the initiation of custom diet. Of note, animal 10 began refusing custom feed pellets around week 16 and was subsequently switched to a normal diet