Eighty-five CKD patients and 44 controls were enrolled from the Cardiology Department of our hospital from July 2023 to January 2024 for the aortic calcification study. This study complied with the Declaration of Helsinki and was approved by the Ethics Committee of Southeast University-affiliated Zhongda Hospital (Number: 2023ZDSYLL298-P01). Blood was collected from the patients along with basic clinical data (Supplementary Table 1). The inclusion criteria were: (1) Willing participants and signing the informed consent form. (2) Individuals aged 18 years or older, excluding pregnant women. (3) Case group: patients with stage III–IV CKD; control group: individuals with normal renal function, no history of hematuria, proteinuria, or CKD, or patients with stage I CKD. The exclusion criteria were: (1) Individuals with a history of ischemic stroke within the past week, intracranial or gastrointestinal hemorrhage within the last 6 months, or major surgery within the previous 30 days. (2) Those with severe respiratory conditions, including but not limited to chronic obstructive pulmonary disease (COPD) or asthma. (3) Severe liver disease (hepatic insufficiency), with aspartate aminotransferase/alanine aminotransferase (AST/ALT) levels exceeding three times the upper limit of normal due to non-cardiac causes, or history of cirrhosis. (4) Severe infectious diseases, or patients with severe dyspnea due to conditions such as pneumonia or COPD. (5) The presence of other severe comorbid conditions that limit life expectancy to less than 6 months. (6) Cardiac conduction disorders, including sick sinus syndrome, advanced atrioventricular block, or bradycardia-induced syncope. (7) Pregnant or lactating women. (8) Participants deemed by the investigator to be otherwise unsuitable for the study.

Eight patients who had undergone aortic replacement were enrolled in this study from the cardiothoracic surgery department of Zhongda Hospital from May 2024 to July 2024 and were included in the immunofluorescence imaging of human aortic tissues. This study complied with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Zhongda Hospital, affiliated with Southeast University (Number: 2024ZDSYLL150-P01). Aortic tissues were collected from the patients. The inclusion criteria were: (1) Patients requiring aortic replacement. (2) Provided written informed consent and voluntarily agreed to take part in the study. (3) Males or nonpregnant females over 18 years of age. The exclusion criteria were: (1) Patients with an autoimmune disease. (2) Pregnant or lactating women. (3) Any other status or condition that would make participation in the study inappropriate.

All animal experiments were approved by the Animal Care and Welfare Committee (Number: 20200326014) and were performed in adherence to the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and the ARRIVE guidelines. All animal procedures were conducted in strict compliance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals (8th Edition, 2011). Efforts were made to reduce the number of animals used and to minimize animal pain and discomfort. Macrophage-specific Adam8 KO mice (Cyagen Biosciences, Suzhou, China) on a C57BL/6 background were generated using the Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated protein 9 (CRISPR/Cas9) system [24] (Supplementary Fig. 1). Mice were housed under controlled conditions at the Southeast University Laboratory Animal Center. They were maintained in a specific-pathogen-free environment within ventilated cages under a 12-hour light/dark cycle, with ad libitum access to food and water. Male mice aged 6–10 weeks were utilized in the current study. The study employed randomization with allocation concealment, and the operators were blinded to group assignment. Mice (n = 28) were fed a 0.25% adenine diet (Macklin, Shanghai, China) for 4 weeks to establish the CKD model. Subsequently, the CKD model mice were switched to a high-phosphorus diet (1.8% phosphorus; Xie Tong Sheng Wu) and received intramuscular injections of calcitriol (1 µg/kg; Sigma-Aldrich, Darmstadt, Germany) twice weekly to induce VC. A control group (n = 12) was maintained on a standard normal diet (ND) group (Xie Tong Sheng Wu, Nanjing, China) and received saline injections in parallel. Tail-tip blood sampling was performed at the start of the experiment (0 W), the fourth week (4 W), and the tenth week (10 W) without anesthesia. Prior to blood collection, the tails were immersed in warm water at approximately 50 °C and disinfected. After blood collection, hemostasis was applied, and the mice were adequately comforted. After 16 weeks, the mice were euthanized via an intraperitoneal injection of sodium pentobarbital (concentration 15 mg/mL, 200 mg/kg), and the aorta and blood were collected. At the conclusion of the experiment, surviving animals were transferred to other approved protocols within the institution for continued use.

Human plasma was processed by centrifugation (1000 rpm, 20 minutes, 4 °C) to pellet cellular debris, and the resulting supernatant was collected for subsequent analysis. Enzyme-linked immunosorbent assays (ELISAs) were performed according to the manufacturer’s instructions (Elabscience Biotechnology Co., Ltd., Wuhan, China). Optical density (OD) was measured at a wavelength of 450 nm.

Following standard protocols, mouse aortas were paraffin-embedded and sectioned. For Von Kossa staining, sections were deparaffinized and rehydrated, then immersed in silver nitrate solution under ultraviolet light for 2 hours. After rinsing in phosphate-buffered saline (PBS), the sections were subjected to hematoxylin and eosin staining. Finally, all sections were dehydrated, mounted, and scanned using a digital slide scanner (Olympus VS200, Tokyo, Japan).

Paraffin sections of human aortic tissue were prepared, dewaxed, and subjected to antigen retrieval. The tissue areas were outlined with a PAP Pen to prevent antibody dispersion. Sections were then incubated with the corresponding primary antibody at 4 °C overnight in a humidified chamber. After washing with PBS on a shaker, a secondary antibody was applied and incubated for 1 hour at room temperature. Nuclei were counterstained with 4^′,6-diamidino-2-phenylindole. Following the application of an anti-fade mounting medium, the slides were coverslipped and imaged with a digital slide scanner (Olympus VS200). For quantitative analysis, five distinct regions of interest per sample were randomly selected at a higher magnification. The mean fluorescence intensity of ADAM8 and $\alpha$-SMA within these regions was measured using ImageJ software (version 2; National Institutes of Health, Bethesda, MD, USA). Briefly, the mean fluorescence intensity was recorded for each of the five selected fields. The average fluorescence intensity per sample was then calculated from these five measurements. These sample averages were subsequently used to compare the differences between the non-calcified group (NG) and the calcified group (CG). Statistical analysis was performed using an unpaired two-tailed Student’s t-test with GraphPad Prism 9.5.0 software (GraphPad Software, Inc. San Diego, CA, USA) and significance was set at p 0.05.

The serum levels of creatinine, phosphorus, urea, calcium, ALP, as well as aortic calcium content, were measured using commercially available assay kits (Elabscience, Wuhan, China). Following the manufacturer’s protocol, aortic tissue segments were lysed in the designated buffer. Calcium ions in the lysate were then chelated with methyl thymol blue under alkaline conditions to form a blue complex. The absorbance of the colored product was measured to quantify calcium content. The results were normalized to the total protein concentration determined with a BCA protein assay kit.

The AAV vector for macrophage-specific Adam8 OE was generated by subcloning the mouse Adam8 coding sequence (NM_007403) downstream of an EGFP-FT2A cassette into a GV650 backbone, under the control of a macrophage-specific F4/80 promoter (sequence provided in Supplementary Information). Briefly, the 3Flag sequence of GV650 was removed to generate a pAAV-F4/80p-EGFP-FT2A-NM_007403-SV40 PolyA construct. The Adam8 fragment was amplified and inserted into the vector via NheI/EcoRI cloning sites. Following plasmid construction, recombinant AAV was packaged, purified, and its titer was determined, and cryopreserved at –80 °C (Genechem Co., Ltd., Shanghai, China). For in vivo delivery, 4 to 6-week-old male mice were administered approximately 8 $\times$ 10¹¹ AAV viral genomes via a tail vein injection. After 4 weeks of AAV expression, the CKD-related VC model was induced in the mice. Animals were euthanized 16 weeks post-AAV injection for downstream analyses.

The aortic tissues were homogenized in RIPA lysis buffer supplemented with protease and phosphatase inhibitors (KeyGen Bio TECH, Nanjing, China). Equal amounts of total protein were loaded onto a pre-cast gel and separated by electrophoresis at a constant voltage of 120 V. Once the bromophenol blue dye front reached the bottom of the gel, proteins were transferred to polyvinylidene membranes using a rapid transfer system at 400 mA for 30 minutes. Following transfer, the membranes were blocked with 5% non-fat milk for 1 hour at room temperature. After blocking, membranes were trimmed and incubated overnight at 4 °C with the indicated primary antibodies. The next day, the membranes were washed and incubated with the corresponding horseradish peroxidase-conjugated secondary antibodies for 1 hour at room temperature with gentle shaking. Protein bands were visualized using an automated chemiluminescence imaging system (Tanon 5200, Shanghai, China). Band intensities were quantified with ImageJ software (version 2; National Institutes of Health, Bethesda, MD, USA), and statistical analysis was performed accordingly. Details of the antibodies used are provided in Supplementary Table 2.

The sample sizes for both the clinical cohort and the mouse experiments were determined based on previously published studies with similar designs. Strict exclusion criteria were applied to the clinical cohort to minimize confounding factors and enhance the robustness of the results. Data analyses were performed using SPSS 26.0 (IBM, Armonk, NY, USA) and GraphPad Prism 9.5.0 software (GraphPad Software, Inc. San Diego, CA, USA).

Continuous data are presented as mean $\pm$ standard deviation. If data met the assumptions of both normality and homogeneity of variance, comparisons between two groups were analyzed using an unpaired t-test, while comparisons among more than two groups were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s or Student–Newman–Keuls post-hoc tests. For data that were normally distributed but had unequal variances, an unpaired t-test with Welch’s correction was used for two-group comparisons, and Welch’s ANOVA was applied for multiple-group comparisons. If data were not normally distributed, the Mann–Whitney test was used for two-group comparisons, and the Kruskal–Wallis test was used for multiple-group comparisons. Categorical variables were analyzed using the chi-squared test. Survival analysis was conducted using the Kaplan–Meier method, and differences between survival curves were evaluated using the log-rank test. A p-value of 0.05 was considered statistically significant.

Aortic calcification was used as the target lesion to explore the relationship between ADAM8 and VC. Ultimately, a total of 129 patients were enrolled for this segment of the research, including 85 CKD patients and 44 controls. Plasma samples were collected, and thoracoabdominal computed tomography (CT) scans were performed on all participants. The demographic and clinical characteristics of patients in both groups are presented in Supplementary Table 1. Patients in the CKD group exhibited higher neutrophil counts (5.21 $\pm$ 2.61 vs. 4.19 $\pm$ 1.29 $\times$ 10⁹/L, p = 0.003), monocyte counts (0.49 $\pm$ 0.24 vs. 0.36 $\pm$ 0.13 $\times$ 10⁹/L, p 0.001), pan-immune-inflammation values (PIVs) (534.39 $\pm$ 622.395 vs. 286.45 $\pm$ 312.03, p = 0.003), systemic immune inflammation index (SII) values (1023.80 $\pm$ 1059.34 vs. 719.79 $\pm$ 569.77, p = 0.036), and plasma ADAM8 levels (942.74 $\pm$ 394.94 vs. 790.26 $\pm$ 237.05 pg/mL, p = 0.007) compared with the control group, along with lower lymphocyte counts (1.15 $\pm$ 0.52 vs. 1.46 $\pm$ 0.52 $\times$ 10⁹/L, p = 0.002) and high-density lipoprotein cholesterol (HDL-C) levels (0.95 $\pm$ 0.39 vs. 1.18 $\pm$ 0.33 mmol/L, p = 0.001). Besides these differences and a reduced red blood cell count, as well as expected variations in renal function parameters, no other laboratory parameters differed significantly between the groups.

The chest and abdominal CT findings were used to categorize the participants into a calcified group (n = 97) and a non-calcified group (n = 32). Patients with aortic calcification were older (67.26 $\pm$ 9.55 vs. 53.91 $\pm$ 16.30 years, p 0.001), had higher ADAM8 levels (1030.98 $\pm$ 488.49 vs. 746.87 $\pm$ 310.93 pg/mL, p = 0.008), blood urea nitrogen (BUN) levels (12.67 $\pm$ 8.73 vs. 8.81 $\pm$ 7.44 mmol/L, p = 0.026), PIV values (494.25 $\pm$ 607.17 vs. 315.14 $\pm$ 278.46, p = 0.025), and SII values (986.23 $\pm$ 1036.20 vs. 719.66 $\pm$ 444.25, p = 0.045) compared to the non-calcified group. Additionally, they had lower lymphocyte numbers (1.17 $\pm$ 0.51 vs. 1.51 $\pm$ 0.56 $\times$ 10⁹/L, p = 0.002), estimated glomerular filtration rates (eGFRs) (48.18 $\pm$ 39.01 vs. 69.83 $\pm$ 41.21 mL/min/1.73 m², p = 0.008), total cholesterol (TC) (3.49 $\pm$ 1.22 vs. 4.18 $\pm$ 1.29 mmol/L, p = 0.007), and low-density lipoprotein cholesterol (LDL-C) levels (1.86 $\pm$ 0.87 vs. 2.23 $\pm$ 0.83 mmol/L, p = 0.037). The remaining parameters were not significantly different between the groups (Table 1).

We collected eight aortic tissue samples from patients who underwent aortic replacement to further analyze the relationship between ADAM8 and VC. The Von Kossa staining results were used to classify the samples into the aortic calcification group (AC) and the non-aortic calcification group (non-AC). The baseline demographic and clinical characteristics of the two groups are presented in Table 2. Patients in the AC group exhibited significantly higher uric acid (UA) levels (270.25 $\pm$ 32.69 vs. 185.75 $\pm$ 56.08 µmol/L, p = 0.040) and significantly lower eGFRs (66.51 $\pm$ 20.34 vs. 103.72 $\pm$ 4.14 mL/min/1.73 m², p = 0.012) compared to the non-AC group, while other indicators were comparable between groups. The results indicated that the AC group had poorer renal function.

Three adjacent sections from the same sample were respectively subjected to Von Kossa staining, immunofluorescence staining for ADAM8 and CD68 and immunofluorescence staining for ADAM8 and $\alpha$-SMA. The Von Kossa staining results were used to categorize the samples into a calcified group (CG) and a non-calcified group (NG). Subsequently, the vascular medial layer was regionally classified based on the results of CD68/ADAM8 immunofluorescence staining and the tissue structure. The CG was further divided into the calcified region, non-calcified region, and macrophage infiltration region, while the NG was partitioned into the non-calcified region and macrophage region.

Through comparative analysis, we observed that the calcified regions were co-localized with ADAM8, demonstrating significant ADAM8 expression (Fig. 1), and the expression level of ADAM8 was positively correlated with the degree of calcification (Supplementary Fig. 2). In the non-calcified regions of the CG, scattered ADAM8 expression was also detected, whereas no ADAM8 expression was observed in the tunica media in the NG (Fig. 1). In the CG, macrophage infiltration was evident around the calcified regions, termed the macrophage infiltration region, predominantly located within the medial layer. This region showed high ADAM8 expression in macrophages. In contrast, the macrophage region in the NG exhibited low ADAM8 expression, with a small number of macrophages, and this region was primarily localized at the junction between the intima and media (Fig. 2).

Fig. 1.

Von Kossa staining images and immunofluorescence staining images of CD68 and ADAM8 in human aortic tissues. ADAM8 expression was co-localized with the calcified region, while hardly any ADAM8 expression was observed in non-calcified regions (n = 4) (p = 0.0038, unpaired t-test). **p 0.01. Abbreviation: ADAM8, A disintegrin and metalloproteinase 8; CD68, Cluster of differentiation 68; DAPI, 4^′,6-Diamidino-2-Phenylindole. Scale bars: 500 µm (overview); 10 µm (Von Kossa); 20 µm (immunofluorescence staining).

Fig. 2.

Von Kossa staining images and CD68 and ADAM8 immunofluorescence staining images in human aortic samples. Comparison of Von Kossa staining and CD68 and ADAM8 immunofluorescence images between macrophage infiltration regions in the calcified group and macrophage regions in the non-calcified group revealed abundant ADAM8 expression in macrophage infiltration regions in the calcified group, while no such phenomenon was observed in the non-calcified group (n = 4) (p = 0.0024, unpaired t-test). **p 0.01. Abbreviations: ADAM8, A disintegrin and metalloproteinase 8; CD68, Cluster of differentiation 68; DAPI, 4^′,6-Diamidino-2-Phenylindole. Scale bars: 500 µm (overview); 10 µm (Von Kossa); 20 µm (immunofluorescence staining).

Immunofluorescence staining for both $\alpha$-SMA and ADAM8 was performed to elucidate the source of ADAM8 in calcified medial tissues. The CG demonstrated a gradual decline in $\alpha$-SMA expression within vascular tissues concomitant with VC development compared with the NG (Fig. 3 and Supplementary Fig. 3). Except for a small amount of ADAM8 and $\alpha$-SMA co-localization around the macrophage infiltration region, a significant amount of ADAM8 was diffusely distributed among cells in the calcified regions, with no co-localization with $\alpha$-SMA. In the NG, abundant $\alpha$-SMA expression was detected in the medial layer, but no ADAM8 expression was observed (Supplementary Fig. 4). The NG predominantly exhibited atherosclerotic plaques, within which substantial ADAM8 expression was detected (Supplementary Fig. 4).

Fig. 3.

Von Kossa staining images and immunofluorescence staining images of $\alpha$-SMA and ADAM8 in human aortic tissues. Comparison of Von Kossa staining and $\alpha$-SMA and ADAM8 immunofluorescence images between macrophage infiltration regions in the calcified group and macrophage regions in the non-calcified group revealed that as the degree of calcification increased, $\alpha$-SMA expression decreased and ADAM8 expression increased (n = 4) (p = 0.0006, unpaired t-test). ***p 0.001. Abbreviations: ADAM8, A disintegrin and metalloproteinase 8; $\alpha$-SMA, alpha-smooth muscle actin; DAPI, 4^′,6-Diamidino-2-Phenylindole. Scale bars: 500 µm (overview); 10 µm (Von Kossa); 20 µm (immunofluorescence staining).

We established macrophage-specific Adam8 KO mice. Knocking out Adam8 did not cause phenotypic changes in the mice. As shown in Fig. 4A, we established an in vivo CKD model using Adam8 KO mice and control mice from the same litter. Knocking out macrophage Adam8 KO did not affect the survival of mice in either sub-model group (Fig. 4B). As renal function deteriorated, the weight of the mice in both model groups decreased equally (Fig. 4C). At the beginning of the experiment (0 weeks), as well as at the fourth and tenth weeks, we evaluated renal function impairment. The results showed that serum phosphorus (Fig. 4D), ALP (Fig. 4E), creatinine (Fig. 4F), and urea (Fig. 4G) levels were significantly elevated in both model groups, with no statistically significant differences observed between the two groups. No significant differences in serum calcium levels were observed among the four groups (Fig. 4H), suggesting that the adenine diet was successful in inducing renal injury and that the KO of macrophage-specific Adam8 did not alter renal function. The aortic calcium content in both CKD groups was increased compared with the ND groups and was higher in the CKD-Flox group than in the CKD-KO group (Fig. 4I).

Fig. 4.

Adam8 KO has no protective effect on renal function in CKD mice. (A) Development of a mouse model for CKD-associated vascular calcification. Created in BioRender. Dong, R. (2025). https://BioRender.com/u16l265. (B) Survival curves. Cumulative mortality in the four groups of mice. ND-Flox mice (n = 6), ND-KO mice (n = 6), CKD-Flox mice (n = 14), and CKD-KO mice (n = 14). (C) Body weights (n = 6–14). (D) Biochemical analyses of serum phosphorus, serum alkaline phosphatase (ALP) (E), creatinine (F), urea (G), and calcium concentrations (H) at 0, 4, and 10 weeks during modeling. (I) Quantification of aortic tissue calcium content (n = 6–14). (p = 0.015, unpaired t-test). Notes: “n value” refers to the number of biological replicates. *p 0.05. Abbreviations: KO, knockout; ND, normal diet. ns, no significant difference.

The extent of VC in the aortic media was assessed by Von Kossa staining and measuring the calcium content. We dissected mice that died during the experiment and compared the calcification of the aortic valve and aorta in the two sub-model groups. Calcification occurred earlier in the aortic valves than in the aortic vessels. Furthermore, the valvular calcification was more severe in the CKD-Flox group compared to the CKD-KO group (Supplementary Fig. 5). The CKD-Flox group displayed more severe vascular calcification than the CKD-KO group (Fig. 5A). Western blots (Fig. 5B and Supplementary Figs. 6,7) showed that Runx2 expression was increased in vascular tissue in the CKD groups (Fig. 5E) Sm22$\alpha$ expression was decreased (Fig. 5C) compared with the ND groups. The CKD-Flox group had higher Runx2 expression and lower Sm22$\alpha$ expression than with CKD-KO group. Given that CKD is an inflammatory disease and that ADAM8 is also related to inflammation, we examined the expression of inflammation-related markers. Interleukin (IL)-18 expression was significantly elevated in the CKD-Flox group (Fig. 5D). This suggests that Adam8 KO may attenuate vascular calcification by suppressing inflammation.

Fig. 5.

Aortic calcification is attenuated by Adam8 KO in CKD mice. (A) Representative Von Kossa staining images of vascular tissue from the four groups (positive staining: black) (n = 6). (B) Western blots of (C) SM22$\alpha$ (p 0.0001 unpaired t-test; original blots/gels are presented in Supplementary Figs. 6,7), (D) Interleukin (IL)-18 (p = 0.0006 unpaired t-test), and (E) Runx2 in aortic tissues of the four groups of mice (p = 0.0009 unpaired t-test). Notes: “n value” refers to the number of biological replicates. **p 0.01, and ***p 0.001. Abbreviations: HE, Hematoxylin and Eosin Staining; GAPDH, Glyceraldehyde-3-Phosphate Dehydrogenase; SM22$\alpha$, smooth muscle protein 22-alpha; Runx2, Runt-related transcription factor 2. Scale bars: 100 µm (overview); 20 µm (high magnification).

The association between macrophage-derived ADAM8 and VC was investigated using macrophage-specific Adam8 OE. This was achieved by injecting the tail veins of 4-week-old Adam8 KO mice with either AAV6-F4/80-Adam8 or a control virus. Following viral delivery, CKD-associated VC was induced in these animals. The results showed that the mice successfully overexpressed ADAM8 after viral injection (Fig. 6B,C and Supplementary Fig. 8). VC was significantly aggravated in the CKD+KO+OE group compared to the CKD+KO+NC group (Fig. 6A). Consistent with this phenotypic enhancement, Western blot analysis revealed a concurrent upregulation of osteogenic (Runx2) and inflammatory (IL-18) markers, and a downregulation of Sm22$\alpha$ in the aortic tissues of the CKD+KO+OE mice (Fig. 6D–G, Supplementary Figs. 9,10).

Fig. 6.

Adam8 OE exacerbates aortic calcification in Adam8 KO mice under CKD conditions. (A) Representative images of Von Kossa staining in the aortic tissues of mice from the two CKD groups (positive staining: black). (B) Western blot analysis and (C) quantification of ADAM8 protein levels (p 0.01, unpaired t-test; original blots/gels are presented in Supplementary Fig. 10). (D) Western blot analysis and quantification of (E) Sm22$\alpha$ protein levels (p 0.001, unpaired t-test; original blots/gels are presented in Supplementary Figs. 8,9); (F) Interleukin (IL)-18 (p = 0.0089, unpaired t-test); and (G) Runx2 in mouse aortic tissues (p = 0.0014, unpaired t-test) (n = 3). *p 0.05, **p 0.01, and ***p 0.001. Notes: “n value” refers to the number of biological replicates. Scale bars: 100 µm (overview); 20 µm (high magnification).