BALB/c IL1rn−/− (CAnN.129P2(MF1)-Il1rntm1Nick/NickH) frozen sperm was obtained from the M.R.C. Harwell Institute, which distributes this strain on behalf of the European Mouse Mutant Archive (EMMA: www.infrafrontier.eu). The repository number is EM:02497. The CAnN.129P2(MF1)-Il1rntm1Nick/NickH mice were originally produced at the University of Sheffield [15,16,17,18]. Specific and opportunistic pathogen-free (SOPF) mice were rederived at Janvier Labs (Le Genest St. Isle, France) by in vitro fertilization using BALB/c IL1rn−/− frozen sperm and oocytes from heterozygous BALB/cAnN IL1n+/− female mice from an initial revitalization. A control group was constituted of 15 wild-type specific pathogen-free (SPF) BALB/cAnN mice (Janvier Labs). As previous studies did not find any difference between males and females regarding the incidence of aortitis in this model [14, 17], we chose to use both in our experiments.

To better refine our data, the second group of 5 BALB/c IL1rn−/− mice, resulting from the mating of the first-generation mice but in an SPF environment in the Centre Universitaire de Ressources Biologiques (CURB, Caen, France), also underwent PET-MR at 9 weeks. These mice were part of a longitudinal study.

All animal procedures were performed according to the European directive 2010/63/EU on protecting animals used for scientific purposes and specific French laws (decree n°2013–118) and were approved by the regional animal ethics committee (Comité d'Éthique NOrmand en Matière d'EXpérimentation Animale, CENOMEXA 054, n°23,523). Mice were housed in an SPF environment in a temperature-controlled room with ad libitum access to standard mouse chow and water. The study is reported in accordance with ARRIVE guidelines.

Immediately after PET-MR, mice were euthanized by decapitation while under deep anaesthesia and unresponsive to all stimuli (5% isoflurane gas in an N2O/O2 mixture (2:1)). All precautions were taken to minimize suffering.

Acquisitions were performed using a hybrid PET-MR 7 T system dedicated to small animals (BioSpec 70/18, Bruker, Ettlingen, Germany). For PET imaging, a static whole-body PET scan (15 min) was obtained. The energy windows were set to 357–664 keV, and the coincidence windows were set to 5 ns. The axial scan length was 117 mm. The image data were corrected for dead time, radioactive decay, attenuation, scatter, and randoms. Attenuation correction was based on segmenting tissue, air, and animal cradles using whole-body 3-dimensional FISP MR acquisition [19]. The sinograms were reconstructed using a 3-dimensional OSEM algorithm (12 iterations, 16 subsets) into a 128 × 128 matrix (slice thickness: 0.75 mm). The MR protocol consisted of a localizer, T2 TurboRARE (TE: 24 ms, TR: 1720 ms, 150  × 150 matrix, FOV: 30 × 30 mm, voxel size: 0.2 × 0.2 × 0.5, RARE factor: 5, averages: 8, slice thickness: 0.5 mm, slices: 30), 2D TOF FLASH MR angiography (TE: 1.657 ms, TR: 12 ms, 125  × 125 matrix, bandwidth: 700 Hz/pixel, flip angle: 80°, averages: 2, slice thickness: 0.6 mm, slices: 60) and whole-body FISP (TE: 2.6 ms, TR: 5.5 ms, flip angle: 10°, image size: 80  × 80, averages: 3, slice thickness: 0.5 mm, slices: 60) sequences. The 2D TOF imaging was gated to the ECG to avoid cardiac motion artefacts, and T2 TurboRARE was gated to the respiratory signal. All 2D images were acquired in axial view. MR images were used as landmarks to quantify [18F]FDG uptake in vessels of interest. However, no qualitative or quantitative data regarding both the morphology and dimension of the aorta could be inferred from these images, because of the size of the vessels and the limited spatial resolution of MR angiography.

To decrease myocardial [18F]FDG uptake, 9-week-old conscious mice were fed a ketogenic diet for two days and then fasted 18 h before the PET-MR scan, and [18F]FDG was intraperitoneally injected (25.9 ± 1.3 MBq, Curium, Glisy, France) [20,21,22]. Fifty minutes later, the mice were anaesthetized with 5% isoflurane gas and maintained with isoflurane 2% gas in an N2O/O2 mixture (2:1) during the PET-MR procedure. MR sequences were obtained first, and PET images were acquired 83 ± 1 min after the [18F]FDG injection.

PET-MR analysis was performed using 3D Slicer 4.11 (http://www.slicer.org/) using the PET-IndiC extension [23,24,25,26]. A rigid transformation (6 degrees of freedom) was applied to align the reconstructed PET to MR images. Volumes of interest (VOIs) were manually drawn with the “Level Tracing Effect” tool in the three axes, encompassing the entire available portion of the ascending thoracic aorta, descending thoracic aorta, abdominal aorta, and inferior vena cava as identified on 2D TOF angiography. To include the vessel wall, the VOIs of the aorta were dilated by the neighbourhood method (8 neighbours). Aortic uptake within each VOI was expressed as the target-to-background ratio (TBR), defined as the ratio of the maximum uptake in the aorta (in Bq/mL) within each VOI of the aorta to the maximum uptake of the residual vascular activity measured in the inferior vena cava.

The aorta was removed, and a piece at the thoracoabdominal junction was taken, weighed, and gamma counted with a designated [18F] protocol for 60 s (Wizard 2470, PerkinElmer, Boston, MA, USA) 126 ± 2 min after the [18F]FDG injection. The results were expressed as counts per minute (CPM) per administered dose and milligram of tissue (CPM.MBq−1.mg−1). As the 5 SPF mice were part of a longitudinal study, these mice have not been euthanized at 9 weeks old and therefore no ex vivo radioactivity measurement was available.

The remaining aorta was embedded in an optimal cutting embedding medium (OCT, Thermo Scientific, Waltham, USA) and stored at -80 °C. Thoracic and abdominal aortas were then longitudinally sectioned (10 µm), mounted on glass slides, and placed on a phosphor imaging plate (BAS-IP Phosphorimaging plate, GE Healthcare Life Sciences, Pittsburgh, PA, USA) 167 ± 4 min after [18F]FDG injection and then imaged with an Amersham Typhoon 5 (GE Healthcare Life Sciences) after overnight exposure and with a pixel resolution of 10 µm. No phosphor imaging autoradiography was performed for the 5 SPF mice, because these mice have not been euthanized at 9 weeks old.

Routine staining was performed using haematoxylin, eosin, and saffron (HES). Immunostaining was performed on a BenchMark XT (Ventana Medical Systems Inc., Tucson, AZ, USA) with the following rabbit antibodies directed against immune cells or endothelial cells and according to the manufacturer’s protocols: anti-CD4 (1:50, Cell Signalling, catalogue number: #25229), anti-CD8α (1:200, Cell Signalling, catalogue number: #98941), anti-CD31 (1:100, Cell Signalling, catalogue number: #77699), anti-CD11c (1:100, Cell Signalling, catalogue number: #97585), anti-CD68 (1:600, Cell Signalling, catalogue number: #97778), anti-IL-17A (1:100, Thermo Fisher Scientific, catalogue number: #PA5-79470), anti-IL-6 (1:100, Thermo Fisher Scientific, catalogue number: #BS-0782R), anti-interferon-γ (1:100, Abcam, catalogue number: #ab216642), anti-CD163 (1:200, Abcam, catalogue number: #ab182422), anti-IL-1β (1:150, Abcam, catalogue number: #ab205924), anti-IL-1α (1:100, Abcam, catalogue number: #ab7632), anti-CD20 (1:50, Abcam, catalogue number: #ab64088). Bound antibodies were detected using avidin–biotin-peroxidase complex (ChromoMap DAB and OmniMap anti-Rabbit HRP, Roche Diagnostics, Meylan, France). Images were acquired using an Olympus VS120 slide scanner (Olympus, Tokyo, Japan) at 20 × magnification. Tissue segmentation was manually performed using the QuPath software package (v0.2.3) [27]. For HES and immunostaining, data were expressed as the mean number of cells per mm2 or as the mean number of positive cells per thousand using the “Cell detection” or the “Positive cell detection” option, respectively (optical density sum, Score compartment: Eosin OD max or DAB OD max with single threshold). Histological staining was not available for the 5 SPF mice, as these mice were part of a longitudinal study.

Quantitative data are expressed as the mean ± standard error of the mean and were analysed using the Mann–Whitney test. Associations were considered significant if the p value was < 0.05. Statistical analyses were performed using GraphPad Prism 7 (GraphPad Software Inc., San Diego, CA, USA). The association between TBR and mouse groups (wild-type mice, first- or second-generation IL1rn−/− mice) or aortic segments (abdominal aorta, ascending or descending thoracic aorta) or aortic segments by mouse group interaction was analysed in a multiple linear regression by the least squares method conducted by JMP 11 (SAS Institute, Cary, NC).