Magnetic resonance imaging and ultrasound elastography in the context of preclinical pharmacological research: significance for the 3R principles

Abstract

The 3Rs principles-reduction, refinement, replacement-are at the core of preclinical research within drug discovery, which still relies to a great extent on the availability of models of disease in animals. Minimizing their distress, reducing their number as well as searching for means to replace them in experimental studies are constant objectives in this area. Due to its non-invasive character in vivo imaging supports these efforts by enabling repeated longitudinal assessments in each animal which serves as its own control, thereby enabling to reduce considerably the animal utilization in the experiments. The repetitive monitoring of pathology progression and the effects of therapy becomes feasible by assessment of quantitative biomarkers. Moreover, imaging has translational prospects by facilitating the comparison of studies performed in small rodents and humans. Also, learnings from the clinic may be potentially back-translated to preclinical settings and therefore contribute to refining animal investigations. By concentrating on activities around the application of magnetic resonance imaging (MRI) and ultrasound elastography to small rodent models of disease, we aim to illustrate how in vivo imaging contributes primarily to reduction and refinement in the context of pharmacological research.

Keywords: 3R principles; in vivo imaging; magnetic resonance imaging; pharmacology; preclinical research; small rodent; translational research; ultrasound elastography.

FIGURE 1

MRI in a post-traumatic OA (PTOA) rat surgical model of cartilage injury. (A) Spin-echo images acquired at 7 T from the same joint, 7 days after surgery. Damaged cartilage in the medial condyle displayed higher signal intensity than in the lateral condyle (arrows). (B) Higher signal intensity translated into increased T₂ relaxation times in damaged cartilage. A reduction of T₂ was observed upon intra-articular administration 1 week after injury onset of a compound aiming to promote cartilage regeneration. Representative safranin-O stained histological images confirmed the beneficial effect of the compound. (C) Quantification of fluorescence applying machine learning tools to immunohistochemistry for collagen-2. See Accart et al. (2022) for details on image acquisition and quantification. ^©The Authors 2022.

FIGURE 2

Sciatic nerve crush in mice. (A) Significantly lower MTR at different regions of the sciatic nerve were determined 1 week post-injury suggesting demyelination. At week 6, MTR was still below baseline after the nerve bifurcation in young mice, while for old mice the region displaying MTR below baseline was larger. (B) Quantitative histological analyses using machine learning tools of myelin basic protein fluorescence signal confirmed lower myelin content in the sciatic nerve as a function of time after the crush. See Giorgetti et al. (2019) for details. ^©The Authors 2019.

FIGURE 3

MRI for the analysis of EAE mice modeling MS: Lesion detection and neurodegeneration. (A) T₁-weighted images acquired at the peak of disease from two EAE animals following intravenous injection of the clinically approved contrast agent Dotarem (Gd-DOTA) as a bolus. Lesions corresponding to leakage of the contrast agent in areas of impaired blood-brain barrier are clearly visible. (B) For volumetric analyses of the whole brain and subareas thereof T₂-weighted images acquired in 12.5 min without administration of contrast material provided sufficient contrast for segmentation. (C) Scheme of study protocol for assessing neurodegeneration in the model. Treatment started at the peak of disease on day 14 after EAE induction. (D) EAE mice treated with fingolimod had improved clinical scores compared to animals receiving vehicle. (E) Neurodegeneration was consistently quantified by MRI in the striatum and cerebellum of EAE mice receiving vehicle but not in those treated with fingolimod. Increased brain derived neurotrophic factor (BDNF) levels were detected in several brain areas and in the serum of fingolimod-treated EAE mice. More details can be found in Smith et al. (2018). ^© 2018 Elsevier B.V.

FIGURE 4

Cuprizone-induced demyelination in the brain of mice. (A) Representative T₂-weighted images from the same mouse acquired before (baseline) and after 5 weeks of cuprizone intoxication. A clear contrast change occurred at the level of the corpus callosum (arrows). Luxol fast blue histology revealed demyelination in the same brain area. (B) Cuprizone induced significant signal increase in T₂-weighted images respectively decrease of magnetization transfer ratio (MTR) in the corpus callosum. (C) Comparison between the MRI signal/MTR parameters and the quantitative histology analysis of luxol fast blue. More details can be found in Beckmann et al. (2018). ^© The Authors. 2018 Open Access.

FIGURE 5

MRI at 4.7 T in the bleomycin model, for animals under spontaneous respiration. (A) Comparison between two-dimensional gradient-echo and UTE images acquired in 22 and 7.4 min, respectively, from one Sprague Dawley rat in the same imaging session at day 15 after bleomycin challenge (4 mg/kg intra-tracheal). The three slices for each acquisition method correspond to the same anatomical location. Note the increase in sensitivity for lesion detection by using ultrashort echo time technique. (B) Detection of bleomycin-induced lung injury by UTE-MRI in a BALB/c mouse. Comparable slices from two-dimensional UTE images (4 min acquisition time, echo time 0.5 ms) before and at different timepoints after oropharyngeal bleomycin administration (1.0 mg/kg/day on 6 consecutive days). Bleomycin-elicited lesions are indicated by the arrows. See Egger et al. (2013), Egger et al. (2014) for more details. ^© 2013 Egger et al. and ^© 2014. The American Physiological Society.

FIGURE 6

Imaging in a mouse lung tumor model. (A) Study protocol scheme. (B) MRI images from one representative mouse at different time points of the treatment phases. Tumor lesions are indicated by the red arrows. (C) Summary of tumor burden based on the quantification of lung lesions in the images. Images from one mouse acquired at baseline (left), at the end of the first (middle) and third cycle of treatment (right). See Egger et al. (2013) for details on image acquisition and lesion quantification. ^© 2013 Egger et al.

FIGURE 7

Imaging analysis of liver injury in mouse models. (A) Longitudinal non-invasive liver fat quantification by MRI. Livers of mice receiving a NASH diet for 20 weeks had an approximately 9x higher fat content than livers of control animals receiving a normal diet. Treatment with a compound (comp) led to significant reduction of liver fat as assessed by MRI, despite continuation of the NASH diet. Biochemical analyses at the end of the study confirmed reduced lipids and triglycerides in NASH animals treated with comp. Of note, despite reducing liver fat, comp had no impact on body weight, pointing to the importance of having a non-invasive readout for liver fat. (B) Liver stiffness assessed by shear wave elastography (SWE). Mice receiving CCl₄ (0.75 μL/g intraperitoneal injection 3x per week) developed significantly higher liver stiffness compared to control animals, which was consistent with biochemical and histological analyses demonstrating higher collagen content and picrosirius red (PSR) staining in the livers of CCl₄-challenged animals. See Gapp et al. (2019) for more details. ^© 2019 The Authors.

FIGURE 8

Microbleeds in the brain of APP23 mice modeling Alzheimer’s disease. (A) Gradient-echo MRI was sensitive to detect microbleeds from an age of 15.5 months onward. Lesion volume was quantified by segmenting the regions displaying signal attenuation. (B) Longitudinal images from two APP23 mice at approximately the same anatomical location for each animal. Treatment started immediately after the acquisition of baseline images at 17.5 months of age. The lesion volume over the three-month-period of treatment is summarized on the right. More details can be found in Beckmann et al. (2016). ^© 2016 The Authors.