MRI-based assessment of endothelial function in mice in vivo

Abstract

While a healthy endothelium serves to maintain vascular haemostasis, a malfunctioning endothelium leads to various cardiovascular diseases, including atherothrombosis. Endothelial dysfunction is characterized by increased vascular permeability, impaired endothelium-dependent responses and various pro-inflammatory and pro-thrombotic changes in endothelial phenotype, all of which could provide the basis for an in vivo diagnosis of endothelial dysfunction. In the present review, we briefly summarize the magnetic resonance imaging (MRI)-based methods available for assessing endothelial function in animal models, especially in mice. These methods are aimed to assess biochemical phenotype using molecular imaging, endothelium-dependent responses or changes in endothelial permeability. All these approaches provide a complementary insight into the endothelial dysfunction in vivo and may offer a unique opportunity to study endothelium-based mechanisms of diseases and endothelial response to treatment.

Introduction

Vascular endothelial cells, representing a structurally and functionally heterogeneous population of cells, form the largest endocrine organ in the body, composed of approximately 1–6 × 10¹³ cells that cover a large surface area. Endothelium maintains vascular haemostasis and regulates vascular tone, permeability of the vessel wall, smooth muscle cell proliferation and migration, thromboresistance, fibrinolysis, and inflammation [1], [2]. Endothelial dysfunction is featured by increased vascular permeability, platelet adhesion, pro-thrombotic response, leukocyte adhesion and pro-inflammatory response [1], [3]. In clinical conditions, a dysfunctional endothelium is diagnosed based on the impairment of nitric oxide (NO)-dependent vasodilation or via pro-thrombotic or pro-inflammatory biochemical markers such as von Willebrand factor (vWf) [4], [5], soluble thrombomodulin (sTM) [5], soluble vascular cell adhesion molecule-1 (sVCAM-1), intercellular adhesion molecule-1 (sICAM-1) [6] or soluble E-selectin [7]. Endothelial dysfunction appears to be a common feature of most if not all cardiovascular diseases, including atherothrombosis, hypertension, diabetes, and heart failure, as well as non-cardiovascular diseases. There is evidence that endothelial dysfunction promotes the development of vascular inflammation, thrombosis, atherosclerotic plaque development and its clinical complications, including myocardial infarct, stroke and peripheral arterial disease [8], [9]. The clinical relevance of the endothelium is supported by the fact that impaired endothelium-dependent vasodilation in the coronary or peripheral circulation of humans has prognostic implications, as it predicts adverse cardiovascular events and poor long-term outcomes [10], [11].

Currently, the assessment of endothelium-dependent changes in vessel diameter is most frequently done based on angiography [12], Doppler ultrasonography [13], plethysmography [14] or tonometry [15]. Since invasive methods have limited usefulness [5], [16], several non-invasive techniques for assessing artery dilation or/and stiffness have been developed, including gold-standard flow mediated dilation (FMD) [17], pulse wave analysis (PWA) [18] and pulse wave velocity analysis (PWV) [19], as well as peripheral reactive hyparaemia (RH-PAT) [20]. Non-invasive techniques based on magnetic resonance imaging (MRI) seem well suited for detecting artery dilation. Indeed, although the number of studies comparing MRI with other techniques is limited, it was demonstrated in healthy volunteers, that the correlation exists between FMD results obtained with the use of standard ultrasound technique and MRI [21], [22], [23]. Authors claim that MRI may be used to provide measures of endothelial function with at least the same accuracy and reproducibility as ultrasound and that variability of FMD response rather than quality of the imaging method may be the limiting factor of the studies [23]. Furthermore, MRI offers a possibility for comprehensive assessment of cardiovascular system taking advantage of diverse contrast sources, full three-dimensional visualization of the vessel, simultaneous measurement of diameter and velocity within one acquisition plane in two aortic locations with the high precision. Comprehensive protocols were developed with the use of MRI-based FMD in combination with MRI-based atheroma characterization [24] aortic stiffness and distensibility [25], [26], [27] as well as PWV [28], [29]. Thus, despite being less available, MRI technology with its versatility and accuracy provides an excellent tool to gain better insight into endothelium-dependent mechanisms of health and disease in clinical research and especially in experimental animal studies.

In particular, mice models of cardiovascular diseases are being increasingly used to increase the understanding of cardiovascular pathologies, and to develop novel phenotype-based methods for pathology progression monitoring in vivo [30], [31]. Accordingly, there are a number of mice models of cardiovascular disease that display the phenotype of endothelial dysfunction, similar to that seen in humans. For example, various strains of genetically modified mice develop spontaneous atherosclerosis [30]. However, studies aimed at analyzing endothelium-dependent vasodilatation or other features of endothelial dysfunction with the use of imaging methods in mice in vivo face greater technical difficulties, as such analysis demands both high spatial and temporal resolution due to the small size of the animals and their rapid heart rhythms [32], [33]. In that context, MRI seems well suited for the purpose and offers advantages over other techniques.

In this paper, we briefly summarize the MRI-based methods for assessing endothelial function in animal models especially in mice, based on biochemical phenotype and molecular imaging, endothelium-dependent vasodilation responses and changes in endothelial permeability. These approaches provide a complementary insight into the endothelial phenotype in vivo.