Introduction

The majority of lethal cardiovascular diseases worldwide occur as a result of atherosclerosis, a maladaptive, non-resolving chronic inflammatory disease that occurs at sites of blood flow disturbance1. While the trigger for atherosclerosis is not yet fully understood, the so called “bad” cholesterol consisting of low-density lipoproteins (LDL) and other triglyceride-rich lipoproteins, is a key factor for the disease development2. Furthermore, macrophages devour the oxidized LDL particles and transform into foam cells, which marks the beginning of the fatty streaks formation3,4. Additionally, the resulting cholesterol crystals co-activate the NACHT, LRR, and PYD domains-containing protein 3 (NLRP3)-inflammasome, a multi-protein complex involved in the activation of interleukin-1β (IL-1β)5. IL-1β is one of the main cytokines in the inflammatory mechanisms of atherosclerosis, activating and mediating the humoral immune response6. Thereby, it does not only activate secondary inflammatory cytokines, like IL-6, but is also responsible for the phenotype transition of vascular smooth muscle cells, responsible for a further increase in IL-1β expression7.

By neutralizing IL-1β in the Canakinumab Anti-inflammatory Thrombosis Study (CANTOS) with the monoclonal Canakinumab antibody, a reduction in recurrent cardiovascular events was observed in patients8,9. Canakinumab binds to IL-1β with high specificity, thus reducing the risk of cross-reactions to other cytokines and leads to the formation of antigen–antibody complexes10. These findings were also validated in a preclinical model of abdominal aortic aneurysm (AAA)11, using the mouse-specific monoclonal antibody 01BSUR which binds to IL-1β and led to a decrease in aortic dilatation and preservation of the extracellular matrix (ECM). Regarding atherosclerosis, mouse studies from recent years demonstrated promising results, i.e. plaque remodeling towards a more stable phenotype12 and decreased vascular inflammation as a result of anti-inflammatory therapy5.

Disturbances in the ECM homeostasis are critically important for plaque stability, migration of inflammatory cells and the retention of lipoproteins, all of which contribute to the progression of atherosclerosis13. Elastin is one of the main ECM components in the vascular wall, responsible for its elasticity and structural integrity14. In the context of atherosclerosis, disrupted elastogenesis15, elastin fragmentation16 and degradation17 promote plaque development. Therefore, characterizing the elastin content of atherosclerotic plaques might offer important insights into plaque composition and vulnerability18. Recently, a low-molecular weight elastin-targeted probe was introduced for molecular magnetic resonance imaging (MRI)19. Various studies demonstrated the successful characterization of plaque burden20 and stability21 in preclinical models of atherosclerosis using elastin-enhanced molecular MRI.

It has been demonstrated previously that IL-1β upregulates elastin gene expression22. We hypothesized that 01BSUR treatment alleviates atherosclerosis and its therapeutic effects can be observed on a molecular level using an elastin-specific MR probe.

Results

Molecular MR imaging for the in vivo assessment of the effects of IL-1β inhibitor 01BSUR treatment

Apolipoprotein E-deficient (ApoE−/−) mice in the control group (n = 10), receiving no treatment with the IL-1β inhibitor 01BSUR, developed visible atherosclerotic plaques in the aortic arch, thoracic aorta, brachiocephalic and carotid arteries after 8 weeks of high-fat diet (HFD) (Fig. 1). Using the elastin-specific probe, elastic fibers within the atherosclerotic plaque could be visualized in vivo. In contrast, the treatment group (n = 10) which received two subcutaneous injections of 01BSUR and 8 weeks of HFD, showed a marked reduction in the development of atherosclerotic plaques compared to the control group, as seen in elastin-enhanced molecular MRI (Fig. 2).

Fig. 1
figure 1

Study design. Two groups of male apolipoprotein E-knockout mice received a high-fat diet. The 01BSUR treatment group (n = 10) received a first subcutaneous injection of the anti-interleukin-1β antibody 01BSUR at the start of the high-fat diet and a second injection 4 weeks later. The control group (n = 10) received no therapy. MR imaging was performed 8 weeks after the start of the high-fat diet. Following the imaging session, animals were sacrificed, and arterial tissue was excised for ex vivo analysis. Created with BioRender.com.

Full size image
Fig. 2
figure 2

Molecular MRI using an elastin-specific probe for visualisation of atherosclerotic plaque burden. Unenhanced T1-weighted MR images showed no relevant signal of the atherosclerotic plaque area. After the intravenous injection of the elastin-specific probe, a clear signal enhancement (with arrows) was observed in the vessel wall corresponding to the atherosclerotic plaque area on T1-weighted images. Ex vivo histological staining confirmed atherosclerotic plaque formation including the expression of elastic fibers within the plaque area. MRI magnetic resonance imaging; EvG Elastica van Gieson; HE hematoxylin–eosin staining; *indicates the arterial lumen; #indicates the plaque area; Scale bars represent 100 µm.

Full size image

T1-weighted MR imaging for the assessment of the elastin-specific probe

In the area of the aortic arch, brachiocephalic and carotid arteries, a low contrast-to-noise ratio (CNR) was shown in both the treatment and control group prior to the intravenous injection of the gadolinium-based elastin-specific probe. T1-weighted imaging after the administration of the elastin-specific probe showed a significant increase (p < 0.05) in CNR in both groups: Mice of the 01BSUR treatment group showed a twofold increase in CNR in the area of the brachiocephalic vessel compared to the unenhanced scan (pre: 3.93042664; post: 8.4007067) whereas in the control group, the MR signal was over 3.5 times higher than prior to the administration of the elastin-specific probe (pre: 3.70679168; post: 13.2982156; Fig. 3A).

Fig. 3
figure 3

Characterization of the plaque-burden by molecular MRI and ex vivo histology in response to anti-inflammatory therapy. (A) A significant increase of contrast-to-noise-ratio was observed after administration of the elastin-specific probe 8 weeks after start of the high-fat diet in both groups (p < 0.05). The control group showed a significantly stronger enhancement compared to the 01BSUR group. (B, C) Mice treated two times with the anti-interleukin-1β antibody 01BSUR showed a significant lower relative plaque size (p < 0.05) as well as lower elastin fiber content within the atherosclerotic plaque area (p < 0.05) compared to mice of the control group. (D) In vivo MRI measurements of the contrast-to-noise area were in good correlation with ex vivo histological analyses (y = 3.25x + 20.80, R2 = 0.77). All data are presented as mean values ± SD.

Full size image

Histological analyses

Analysis of plaque size and elastin amount by Elastica van Giesson (EvG) staining

To evaluate the effect of the 01BSUR treatment on atherosclerotic plaque formation, Elastica van Giesson (EvG) staining was performed to assess the plaque size and the amount of elastic fibers in the plaque area. Mice of the 01BSUR treatment group showed a significant reduction in plaque formation and relative size (p < 0.05) compared to mice of the control group (01BSUR group: 4.9%, control group: 16.5%, Fig. 3B). Moreover, EvG staining demonstrated a significant decrease in the amount of elastin in the plaque area of the 01BSUR treatment group (42.86% ± 16.66%) compared to the control group (72.30% ± 6.87%, p < 0.05, Fig. 3C).

Histological measurements of elastin content were in good correlation with the in vivo MR measurements of signal enhancement after administration of the elastin-specific probe (y = 3.25x − 20.80 R2 = 0.77; Fig. 3D).

Analysis of inflammation of the atherosclerotic arterial wall by immunofluorescence

For evaluation of the inflammatory processes within the atherosclerotic plaque, different immunofluorescence (IF) stainings were performed. The presence of CD68 + macrophages within the plaque area was evaluated and demonstrated a fourfold decrease (Fig. 4A) in the 01BSUR treatment group (1.568%) compared to the control mice (6.951%), indicating a significant (p < 0.5) reduction of inflammation within the atherosclerotic arterial wall by 01BSUR treatment. Importantly, IF staining against IL-1β demonstrated five times lower interleukin-1β amount within the plaque area in 01BSUR treated mice compared to the non-treated control group (Fig. 4B). Furthermore, a clear colocalization of areas positive for IL-1β and CD68 was observed (Fig. 4C). Additionally, a strong correlation between the absolute amount of interleukin-1β and CD68 within the plaque area was measured (y = 1.22x + 0.35, R2 = 0.87, Fig. 4D).

Fig. 4
figure 4

Ex vivo histological analyses of inflammatory markers. (A, B) Mice treated two times with 01BSUR showed a significant lower expression of CD68 + macrophages (p < 0.05) and interleukin-1β (p < 0.05) in atherosclerotic plaque area after 8 weeks of high-fat diet compared to mice of the control group. (C) Immunofluorescence analyses: Macrophages (CD68 +) are shown in green, interleukin-1β in red and cell nuclei (DAPI) in blue. The stacked image shows a co-localization of interleukin-1β with macrophages. (D) The amount of CD68 + macrophages and IL-1β-expression within the atherosclerotic plaque area were in good correlation (y = 1.22x + 0.35; R2 = 0.87). *indicates the arterial lumen; #indicates the plaque area; Scale bars represent 100 µm.

Full size image

Spatial distribution of the gadolinium-based elastin-specific probe by laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS)

For visualization and spatial localization of the gadolinium-based elastin-specific probe in the atherosclerotic arterial wall, LA-ICP-MS was performed (n = 1). On histological sections of the brachiocephalic artery a strong co-localization of targeted gadolinium with elastic fibers was demonstrated (Fig. 5).

Fig. 5
figure 5

Laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) for spatial localization of gadolinium in the atherosclerotic plaque area. (A) Unenhanced light microscopy image of an animal (n = 1) from the control group. (B) Corresponding image of the LA-ICP-MS visualizing the gadolinium-based elastin-specific probe within atherosclerotic plaque by the gadolinium signal. (C) Elastica van Gieson staining revealed a clear colocalization of gadolinium-accumulation with elastic fibers in the atherosclerotic plaque. *indicates the arterial lumen; #indicates the plaque area; Scale bars represent 100 µm.

Full size image

Discussion

In the current study, we demonstrated the feasibility of molecular MRI using an elastin-specific probe to visualize and evaluate an IL-1β targeted anti-inflammatory therapy against atherosclerosis in a murine model.

The role of interleukin-1β in the development of cardiovascular disease

Through binding and neutralizing the cytokine IL-1β, the antibody 01BSUR decreases the pro-inflammatory and tissue-degrading activities of interleukin which results in a marked reduction of the development and progression of a variety of acute and chronic inflammatory diseases. In the case of atherosclerosis, IL-1β promotes the exacerbation of established atherogenesis23,24. In the present study, ex vivo analyses revealed a significantly lower IL-1β expression within the atherosclerotic lesion in 01BSUR-treated mice compared to the non-treated control group. These findings underline the critical pro-inflammatory role of interleukin-1β as one of the key mediators of inflammation during the formation and progression of atherosclerotic plaques. Produced by intrinsic vascular wall cells and lesional leukocytes, IL-1β induces procoagulant activity and promotes especially the adhesion of monocytes and leukocytes to vascular endothelial cells, as well as the growth of vascular smooth muscle cells, thus initiating plaque development, and influencing plaque stability1,8,25. Our results are in line with Kirii et al.26, demonstrating that a genetic deletion of IL-1β in ApoE−/− mice reduces atherosclerotic lesions by 33%, whereas the deficiency of IL-1 receptor antagonist (IL-1Ra) significantly increased the atherosclerotic lesions formed27. Furthermore, chronic treatment with IL-1β in vivo caused coronary intimal lesions and vasospastic responses in porcine coronary arteries28.

Collectively, these studies suggest that interfering with IL-1β signaling in the vascular wall, adopting strategies such as upregulation of the activity of IL-1Ra, downregulation of IL-1 production in the vascular wall or effecting IL-1 receptor signaling, can potentially be used as a therapeutic target in atherosclerosis25. With this study, we aimed to visualize those effects using elastin-enhanced molecular MRI.

Anti-inflammatory therapies in atherosclerosis

In recent years, several different inflammatory pathways have been identified in the initiation and progression of atherosclerosis. Following the CANTOS study, in which the effects of IL-1β antibody were evaluated in patients after myocardial infarction8, several studies have investigated the use of 01BSUR against atherosclerosis in preclinical models and produced conflicting results. Here, we found a significant reduction in plaque size, elastin and macrophage content in the treatment group compared to the controls. In a study by Dragoljevic et al., 01BSUR treatment led to a reduction in macrophage and lipid content, yet an increase in collagen content, speculating that those changes lead to plaque stabilization12. Hettwer et al. also reported positive effects of 01BSUR, reflected in reduced inflammatory markers, total plaque size and smaller necrotic cores5. On the contrary, Gomez et al. argue that endogenous IL-1β has atheroprotective effects, since treatment with 01BSUR led to a decreased smooth muscle cell and collagen content and increased macrophage content in the fibrous cap, while no reduction of plaque size was seen29. It must be pointed out that the abovementioned studies used different mouse strains, atherosclerosis induction techniques and treatment regimens, which might explain the contradictory findings.

A large-scale clinical trial demonstrated that a low dose of colchicine significantly lowers the risk of ischemic cardiovascular events following a recent myocardial infarction30. Colchicine inhibits tubulin polymerization30,31 and limits the NLRP3 inflammasome activity, consequently suppressing IL-1β activity32. In a rabbit model of atherosclerosis, colchicine treatment resulted in plaque stabilization by reducing the plaque burden and inflammatory activity without a marked macrophage reduction33. This is partially in line with the results from our mouse model under 01BSUR therapy, where we found a reduction in plaque size and a statistically significant reduction of CD68 + cells in the vascular wall.

Another common target for cardiovascular trials is IL-6. Tocilizumab, an IL-6 antibody has been shown to improve the clinical outcome for patients with myocardial infarction by rapidly modulating neutrophile function34,35. Other drug targets such as methotrexate, an antimetabolite and immunosuppressant, did not lead to a significant reduction of pro-atherosclerotic markers, including IL-1β36.

Effects of IL-1β on elastin metabolism in cardiovascular disease

Previous studies have demonstrated a relationship between IL-1β and elastin metabolism in the development of cardiovascular disease. In animal models of aneurysms and aortic dissections, disease induction led to elevated IL-1β tissue levels, and concurrently, breakage and reduction in elastic fiber content37,38. Conversely, treatment with an IL-1β antibody or genetic depletion of IL-1β in these animal models led to preservation of elastin fiber levels and morphology37,39. Regarding atherosclerosis, ApoE−/− IL-1Ra−/− mice showed more pronounced signs of arterial inflammation and destruction of the elastic laminae during plaque development compared to ApoE−/− mice40. Taken together, these results suggest that with progressing atherosclerosis, an increase in local IL-1β and decrease in elastin occurs. In contradiction with those results, we found that 01BSUR therapy resulted in lower elastin content in atherosclerotic plaques compared to non-treated animals. A possible explanation for this contradiction is the dynamic interplay between elastolysis and elastogenesis. It has been demonstrated that IL-1β upregulates elastin gene expression22 and ineffective elastin synthesis takes place during atherosclerosis progression15. In turn, this can lead to an increase in total elastin content20,21 in advanced disease stages. The resulting immature elastin, in combination with destructed elastic fiber fragments feeds into a vicious cycle of pathological elastin remodeling41.

Therapy monitoring using molecular elastin-specific MR imaging

So far, the elastin-specific probe used in this study has been successfully implemented for therapy monitoring in several preclinical models. Here, we showed its efficacy for non-invasive detection of the therapeutic effects of a murine IL-1β antibody on a molecular level. These results are in line with data from an abdominal aortic aneurysm (AAA) model, where 01BSUR therapy prevented extensive aortic dilatation and elastin destruction. These findings were confirmed with elastin-enhanced MRI11. Similarly, Makowski et al. found a decrease of CNR and relative atheroma size following statin therapy in an ApoE−/− mouse model of progressive atherosclerosis20. Beyond cardiovascular disease, elastin-enhanced MR imaging allowed for longitudinal monitoring of the therapeutic effects of an anti-fibrotic drug42 in kidney fibrosis. Moreover, the beneficial effects of non-invasive elastin imaging go beyond pharmacological therapies. Following thermal ablation of liver cancer, it is possible to quantitively asses the ablation-induced ECM remodeling in the periablational rim43.

Translational potential and limitations

The use of a 3 T clinical MRI scanner instead of a high-field preclinical device in this study represents a key factor for future translation into patients. Moreover, the drug dosages and administration routes for therapy and imaging were based on the clinical guidelines for Canakinumab8 and Gd-based contrast agents, respectively20. The size, molecular composition and blood half-life of the elastin-specific probe are comparable to that of clinically approved MRI contrast agents20.

While we yielded some promising results, our study also has several limitations. First, there are some important dissimilarities regarding atherosclerosis development between ApoE−/− mice and humans. ApoE−/− mice develop accelerated atherosclerosis, with initial lesions being visible as early as 6 weeks of age44. In our study, the first dose of 01BSUR was administered at 8 weeks of age, shortly after initial plaque formation, in parallel with the high fat diet start. In the future, it would be beneficial to test the efficacy of 01BSUR on late-stage atherosclerotic plaques. Atheroma development in humans starts later in life and the progression spans over several decades45. Additionally, the predominant type of plasma cholesterol in hyperlipidemic ApoE−/− mice is very low-density lipoproteins (VLDL), compared to low-density lipoproteins (LDL) in humans46,47. Future studies focusing on plaque composition on the molecular level and associated risk factors in both animal models and human cohorts are needed to establish the significance of regional atheroma morphology for cardiovascular events.

Second, only male mice were examined in this study. Previous literature reports differences in atheroma development48 and response to environmental triggers49 in male and female atherosclerotic ApoE−/− mice. Adding a female subgroup in future experiments would allow us to study sex-specific therapy response differences.

To conclude, we demonstrated that 01BSUR hinders the progression of atherosclerosis by reducing plaque size, inflammatory markers, and elastin content. Using an elastin-specific molecular probe, we could non-invasively quantify the therapeutic success of an anti-IL-1β by signal intensity measurements. In the future, molecular MRI techniques may support the diagnosis and treatment of atherosclerosis by characterization of plaque composition and non-invasive monitoring of pharmacological therapies in patients.

Methods

Animal experiments

All animal experiments were performed with respect to the guidelines and regulations of the Federation of Laboratory Animal Science Associations (FELASA) and the local Guidelines and Provisions for Implementation of the Animal Welfare Act and were approved by the Landesamt für Gesundheit und Soziales (LaGeSo) Berlin. The study is reported in accordance with the ARRIVE guidelines for reporting animal research when applicable. Randomization was performed when allocating the animals to each group. Blinding was not possible during the in vivo experiments and was not performed during data analysis.

The mice used in this study were obtained from the Research Institute of Experimental Medicine at the Charité–Universitätsmedizin Berlin. Maintenance of the animals included barrier conditions, a dark/light cycle of 12 h, an ambient temperature of 20 °C, and a humidity of 45%. Two groups of apolipoprotein E-knockout (B6.129P2-ApoEtm1Unc/J) mice (male, 8-week-old, n = 10 per group) were used. Both groups were fed a HFD containing 22% fat (C 1090-45; Altromin Spezialfutter GmbH, Lage, Germany) for 8 weeks, starting at the age of 8 weeks. The diet as well as water were provided ad libitum. To evaluate the therapeutic effect of the anti-IL-1β antibody 01BSUR on atherosclerotic plaque formation, one group (01BSUR treatment group, n = 10) received subcutaneous injections of 01BSUR (Novartis, Basel, Switzerland, 0.3 mg/kg) twice: the first injection at the start of the HFD and the second injection 4 weeks after the start of the HFD. The second group of mice (n = 10), received no therapy, serving as the control group (Fig. 1).

MRI was performed 8 weeks after the start of the HFD. First, mice were anaesthetized using a combination of Medetomidine (500 µg/kg), Fentanyl (50 µg/kg), and Midazolam (5 mg/kg) which were administered intraperitoneally. Second, unenhanced MR imaging was performed, followed by the intravenous injection of the elastin-specific probe (Lantheus Medical Imaging, North Billerica, Massachusetts, USA; 0.2 mmol/kg) and further imaging.

Following the imaging session, animals were sacrificed by cervical dislocation and exsanguinated by arterial perfusion of saline. The brachiocephalic and carotid arteries, aortic arch and thoracic aorta were excised for ex vivo analysis.

Interleukin-1β inhibitor 01BSUR

In this study, mice in the treatment group (n = 10) received the IL-1β inhibitor 01BSUR (Novartis, Basel, Switzerland) two times: the first injection (s.c.) at the start of the HFD and the second s.c. injection 4 weeks after the start of the HFD. A dosage of 0.3 mg per mouse per administration was used (0.6 mg per mouse in total), equivalent to the clinically approved dose of Canakinumab used in the CANTOS study8. 01BSUR represents an IgG2a/k monoclonal mouse antibody that binds with high specificity to IL-1β with a dissociation constant of (KD) = 302 pM and a half-life of 14 days50. Due to its anti-inflammatory effects, the antibody has already been evaluated in various other models of diseases, including aortic aneurysms50 as well as aortic calcification51 and collagen-induced arthritis52.

Gadolinium-based elastin-specific contrast probe

For visualization and characterization of the arterial wall, we used an elastin-specific probe (Lantheus Medical Imaging, North Billerica, Massachusetts, USA) based on gadolinium). This probe has already been studied for the evaluation of atherosclerosis progression53 and AAAs in mice54,55,56. With a molecular weight of 856 g/mol, a short blood half-life and a predominant clearance by the renal system, the probe is highly comparable to contrast agents already used in clinical practice20. Following intravenous injection, the highest uptake within the aorta was shown after 30 to 45 min (13.2 ± 2.3% ID g−1)20.

In vivo MR imaging

Mice in the treatment (n = 10) and control group (n = 10) were imaged after 8 weeks of HFD. For MR imaging, a clinical 3 T MR Scanner (Biograph, Siemens Healthcare Solutions, Erlangen, Germany) in combination with a clinical single loop coil (47 mm, Siemens Healthcare Solutions, Erlangen, Germany) was used. The anesthetized mice received a venous access via the tail vein for administration of the contrast probe and were then placed in a prone position on the coil. During the image session, the body temperature (37 °C) of the animals was constantly monitored with an MR-compatible heating system (Model 1025, SA Instruments Inc, Stony Brook, NY).

MR imaging was performed as previously described53:

For an anatomical overview and localization of the thoracic aorta, carotid and brachiocephalic arteries, low-resolution three-dimensional gradient echo scout scan was performed in sagittal, coronal and transverse orientation using the following parameters: field-of-view (FOV) = 280 mm, matrix = 320, slice thickness = 3 mm, TR/TE = 7.7/3.7 ms, flip angle = 20° and slices = 10. The scout scan was followed by a two-dimensional (2D) time-of-flight (TOF) scan in transverse orientation for visualization of the aortic arch and the brachiocephalic artery. Imaging parameters included: FOV = 200 mm, matrix = 960, in plane spatial resolution = 0.2 × 0.2 mm, slice thickness = 500 µm, TR/TE = 35/4.5 ms, flip angle = 90° and slices = 26. From the TOF dataset, a maximum intensity projection (MIP) was generated for display of an arterial angiogram of the aortic arch, the brachiocephalic and carotid arteries to plan the subsequent contrast-enhanced sequences. The inversion recovery scan to visualize the G-based contrast agent was preceded by a 2D Look-Locker sequence planned perpendicular to the ascending aorta, which was used to determine the optimal inversion time (TI) for blood signal nulling. Imaging parameters of the Lock Locker sequence included: FOV = 300 mm, matrix = 750, in plane spatial resolution = 0.4 × 0.4 mm, slice thickness = 1.5 mm, TR between subsequent IR pulses = 1000 ms, and flip angle = 15°. Imaging parameters of the high-resolution 3D inversion recovery gradient echo late Gd enhancement (LGE) sequence scan employed for visualization of G-based molecular probe were: FOV = 57 mm, matrix = 416, in plane spatial resolution = 0.137 × 0.137 mm, slice thickness = 370 µm, slices = 56, TR/TE = 12.1/5.7 ms, TR between subsequent IR pulses = 1000 ms, and flip angle = 30°.

Assessment of magnetic resonance imaging signal

Quantification of the MRI signal and morphometric measurements were conducted on high–resolution MR images as previously described54.

Using OsiriX (version 7.1, OsiriX foundation), Regions Of Interest (ROIs) were co–localized with the atherosclerotic plaque (highest signal within the arterial wall) and defined as areas of enhancement. The CNR was then calculated: CNR = (Combined vessel wall and atherosclerotic-plaque-signal − Blood signal)/Noise. The noise was defined as the standard deviation in pixel intensity from a ROI placed in the background air anterior to the brachiocephalic artery.

Histological analysis of the atherosclerotic plaques

For histological analysis, arterial tissue, including the brachiocephalic and carotid arteries, aortic arch and thoracic aorta, was frozen at − 20 °C following excision. Then tissues were embedded in optimal cutting temperature (OCT) compound (Sakura Finetek, Breisgau, Germany) and subsequently cut into 10 μm cryosections. Samples were mounted on SuperFrost Plus adhesion slides (Thermo Fisher Scientific, Waltham, USA) and stained using Miller’s Elastica van Gieson stain (EvG) and Hematoxylin and Eosin (HE).

For further morphometric analysis and elastin quantification, slides were examined and photographed using a digital light microscope (BZ-X800, Keyence, Osaka, Japan) and corresponding Keyence BZ-X800 Analyzer software (Keyence, Osaka, Japan). The %EvG stain area per vessel cross section was determined as previously described54,57:

First, a reference standard was set by marking the color profile of the normal elastic laminae in the medial wall. Any structure with this specific color profile within the adventitia was automatically segmented and the resulting areas were documented. We then divided the segmented area by the overall vessel surface and thus determined the %EvG stain per vessel cross-section.

Immunofluorescence analysis

For the visualization and subsequent quantification of IL-1β and CD68 + macrophages within the arterial wall, immunofluorescence (IF) staining was performed. Cryosections were obtained as described above. First, antibodies were diluted in Dako REAL Antibody Diluent (Dako, Glostrup, Denmark). Second, sections were incubated with a polyclonal IL-1β antibody (rabbit anti-mouse IL-1 beta, 1:100 dilution, Bio-Rad, Hercules, USA) or a monoclonal CD68 antibody (rat anti-mouse MCA1957, 1:100 dilution, BioRad, Hercules, USA) overnight at 4 °C. After a wash using PBST (0.05% Tween20, pH 7.4), slides were incubated with polyclonal secondary antibody labeled with AlexaFluor 568 (donkey anti-rabbit IgG, Thermo Fisher Scientific, Germany, 1:200 dilution or goat anti-rat A11077, 1:200 dilution, Invitrogen, Waltham, USA) for 1 h at room temperature. Subsequent counterstaining and mounting were performed with Roti®-Mount FluorCare (Carl Roth, Germany).

For interleukin and CD68 + quantification, slides were examined and photographed using a digital light microscope (BZ-X800, Keyence, Osaka, Japan). Subsequent morphometric analysis was performed using Keyence BZ-X800 Analyzer software (Keyence, Osaka, Japan).

Element-specific bioimaging using laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS)

LA-ICP-MS was performed as described previously54. Arterial tissue was cut into 10 µm cryosections at − 20 °C and mounted on SuperFrost Plus adhesion slides (Thermo Fisher Scientific, Waltham, MA, USA). The LA-ICP-MS analysis was performed using an LSX 213 G2 + laser system (CETAC Technologies, Omaha, USA) which was equipped with a two-volume HelEx II cell connected via Tygon tubing to an ICPMS-2030 (Shimadzu, Kyoto, Japan). With a spot size of 15 µm samples were ablated via line-by-line scan using a scan speed of 45 µm/s and 800 mL/min He as transport gas. The analysis was performed in collision gas mode with He as collision gas, with 50 ms integration time for the analyzed isotopes 31P, 57Fe, and 64Zn and 75 ms integration time for the two Gd isotopes 158Gd and 160Gd. Matrix-matched standards based on gelatin were used for the quantification of Gd. Nine gelatin standards (10% w/w) including a blank, were spiked with different Gd concentrations (1 to 500 µg/g). Averaged intensities of the scanned lines of the standards were in good linear correlation with a regression coefficient R2 = 0.991 within this concentration range. Limit of detection (LOD) and limit of quantification (LOQ), calculated with the 3σ- and 10σ-criteria, were 21 ng/g and 71 ng/g Gd. The quantification and visualization were performed using in-house developed software (WWU Münster, Münster, Germany).

Statistical analysis

The statistical analysis was performed using Origin 2022 (OriginLab Corporation, Northhampton, USA). Values are specified as mean ± standard deviation. Normal distribution was examined via Shapiro–Wilk test. Statistical comparison of groups was performed by analysis of variance (ANOVA) and Sheffe’s test. An unpaired, two-tailed Student’s t-test was applied for the comparison of continuous variables. Linear regression was applied to determine the relationship between in vivo and the ex vivo measurements. A p-value < 0.05 was considered to indicate a statistically significant difference.