TNALP associated calcium phosphate microdeposition contributes to thoracic aortic dissection susceptibility

Abstract

Aortic dissection is a life-threatening cardiovascular emergency characterized by rapid progression and high mortality. Although medial degeneration is a key pathological feature of aortic dissection, the mechanisms that drive aortic wall instability remain incompletely understood. Aortic calcification is a pathological mineralization process that can alter vascular structure and function, but its role in aortic dissection has received limited attention. Here we show that tissue nonspecific alkaline phosphatase associated calcium phosphate microdeposition is linked to aortic medial degeneration and dissection susceptibility. Human aortic dissection tissues and diseased mouse aortas show increased mineralization together with elevated tissue nonspecific alkaline phosphatase activity. Calcium phosphate deposits promote vascular smooth muscle cell phenotypic switching and inflammatory remodeling, while tissue nonspecific alkaline phosphatase facilitates mineral deposition under permissive conditions. Reducing tissue nonspecific alkaline phosphatase activity or expression attenuates aortic calcium phosphate deposition, medial degeneration and dissection in mice. These findings identify tissue nonspecific alkaline phosphatase associated microdeposition as a mechanism contributing to aortic wall vulnerability.

This study investigates the role of tissue nonspecific alkaline phosphatase associated calcium phosphate microdeposition in aortic dissection. It shows that this process promotes aortic medial degeneration and increases dissection susceptibility.

Introduction

Aortic dissection (AD) is a life-threatening cardiovascular emergency characterized by blood entering through an intimal tear and dissecting along the aortic wall, resulting in a separation between the intimal and medial layers^1,2. AD is frequently associated with branch artery malperfusion, persistent hypertension, and intense pain; once rupture occurs, the mortality rate can be as high as 90%. Therefore, timely assessment and surgical intervention remain the mainstay of clinical management for AD^3,4. However, due to its asymptomatic nature before onset and rapid progression once it occurs, early prediction and prevention of AD remain highly challenging.

Current guidelines primarily focus on the management of established risk factors for AD, such as genetic connective tissue disorders (e.g., Marfan syndrome), hypertension, and smoking, through lifestyle modifications and pharmacological interventions^4,5. However, these risk factors do not fully explain the underlying mechanisms of AD, which remain incompletely understood. It is widely accepted that aortic medial degeneration constitutes a pathological hallmark of AD, characterized by the loss of the contractile phenotype of vascular smooth muscle cells (VSMCs), degradation of elastic fibers, and accumulation of extracellular matrix components^6,7. Of interest, our previous propensity score-matched analysis involving 125 patients with type B AD and healthy controls identified thoracic aortic calcification volume and plasma tissue-nonspecific alkaline phosphatase (TNALP) levels as independent risk factors for type B aortic dissection (TBAD)⁸. Furthermore, Mendelian randomization analysis based on single nucleotide polymorphisms (SNPs) provided causal evidence supporting the role of elevated plasma TNALP in promoting the risk of AD⁸. These findings suggested that aortic calcification (AC) may contribute to the risk of AD.

AC is a detrimental pathology characterized by the abnormal deposition of calcium-phosphate crystals within the aortic wall, commonly observed in aging, diabetes, atherosclerosis, and chronic kidney disease^9–12. AC has been reported to be independently associated with cardiovascular events and mortality. In the aorta, AC not only impairs elastic function, leading to increased aortic stiffness, but also contributes to structural and functional abnormalities of the aortic wall^13,14. The transition of VSMCs from a contractile to an osteogenic phenotype is considered a key mechanism underlying “active aortic calcification.”^15,16 During this process, cytosolic Ca²⁺ and phosphate (Pi) incorporate with TNALP into exosomes that bud off the plasma membrane and bind to matrix proteins such as collagen. Crystals in the exosomes finally grow into nano-sized hydroxyapatite (nHAp) [Ca10(OH)2(PO4)6]¹⁷. TNALP hydrolyzes pyrophosphate (an inhibitor of calcification) to produce phosphate, thereby promoting the formation of hydroxyapatite crystals. Consequently, increasing attention has been paid to TNALP as a potential therapeutic target for inhibiting AC^18–21. While the role of TNALP in vascular calcification has been relatively well established, its involvement in AD remains unexplored.

In this study, building on our previous clinical evidence linking aortic mineralization and plasma TNALP to AD risk, we hypothesized that TNALP-driven calcium-phosphate microdeposition contributes to aortic medial degeneration and thereby increases susceptibility to AD.

Results

TNALP upregulation and calcium-phosphate microdeposition are associated with human AD tissues

Our previous study showed a significant increase in aortic calcium volume burden and TNALP in AD patients⁸. To validate calcification phenotype in human AD, we collected thoracic aortic tissues from five patients with dissection undergoing open aortic replacement surgery and aortas from five control subjects who underwent coronary artery bypass grafting patients without dissection. The two groups’ characteristics were presented in Supplementary Table 1. Notably, plasma TNALP level was higher in AD patients than controls (81.00 ± 6.76 vs. 63.80 ± 3.85 U/L, P = 0.056), consistent with our previous findings⁸. To assess calcification and collagen deposition (a pathological hallmark of dissection)²², aortic tissues were subjected to von Kossa staining and Masson’s trichrome staining. EVG staining revealed fragmented and disorganized elastic lamellae in AD patients, while von Kossa staining demonstrated significantly enhanced calcification accompanied by increased collagen deposition in aortic tissues (Fig. 1A, B). Consistent with previous findings²³, our histological analysis of AD aorta revealed a higher prevalence of microcalcifications (granular or punctate, <50 μm). ¹⁸F-NaF PET-CT is an effective modality for detecting microcalcifications, and the radiotracer ¹⁸F-NaF has been clinically validated for identifying microcalcifications in human^24,25. Therefore, we employed ¹⁸F-NaF autoradiography to detect microcalcifications, observing distinct ¹⁸F-NaF uptake in AD aorta (Fig. 1C). As a marker of VSMCs osteogenic differentiation and the key executor of calcium-phosphate deposition²⁶, TNALP was found to be upregulated in both plasma (Supplementary Table 1) and thoracic aortic tissues of AD patients (Fig. 1D, E). Taken together, these results indicate enhanced aorta mineralization in AD patients.

Fig. 1. TNALP upregulation and calcium-phosphate microdeposition are associated with human AD tissues.

A Representative elastin Van Gieson (EVG) staining, von Kossa staining and Masson staining of aortic dissection (AD) samples and non-AD samples from human subjects. The leftmost column represents non-AD aortic samples, while the right columns correspond to aortic tissues from AD patients 1, 2, and 3, respectively. The black arrows indicate the microcalcification/calcification. Scale bar = 50 μm. B Statistical analysis of von Kossa staining and Masson staining. C Representative EVG staining and ¹⁸F-NaF autoradiography images of aortic tissues from two AD patients. The color gradient in the autoradiographic images reflected the intensity of ¹⁸F-NaF uptake. Scale bar = 1 mm. D Representative elastin Van Gieson (EVG) staining and TNALP immunohistochemistry AD samples and non-AD samples from human subjects. Scale bar = 200 μm. E Statistical analysis of TNALP immunohistochemistry. Data presentation: data in (B, E) are presented as mean ± SEM. Each data point represents an individual subject as a biological replicate with similar results. The sample sizes per group for (B) were non-AD (n = 5), AD (n = 5). The sample sizes per group for (E) were non-AD (n = 4), AD (n = 4). Exact n is provided in the Supplementary Data file. Statistical analysis: statistical analysis of (B) was performed using two-sided unpaired Student’s T test. Statistical analysis of (E) was performed using two-sided Mann–Whitney test. Source data are provided as a Supplementary Data file. EVG elastin Van Gieson, AD aortic dissection, TNALP tissue-nonspecific alkaline phosphatase.

BAPN-induced TAAD is accompanied by increased aortic TNALP activity and ultrastructural calcium-phosphate deposition

We further determined TNALP and aorta mineralization in the mouse model of thoracic aortic aneurysm and dissection (TAAD) in the C57BL/6J induced with 0.3% BAPN for 14 days (early stage) or 28 days (endpoint). EVG staining confirmed the presence of a false lumen in the mouse aorta (Fig. 2A). Consistent with the observations in humans, plasma TNALP levels were significantly elevated in BAPN-induced mice at both 14 and 28 days compared to controls (14-day: 397.0 ± 13.3 vs. 284.9 ± 5.4 U/L; 28-day: 202.7 ± 9.7 vs. 153.1 ± 10.9 U/L) (Fig. 2B). In the thoracic aorta of BAPN-induced mice, both TNALP expression and activity were significantly elevated at the 14-day time point (Fig. 2C, D, and Supplementary Fig. 1A, B). IHC and IF staining confirmed persistent upregulation of TNALP expression in BAPN-induced mice at 28 days (established AD phase) (Supplementary Fig. 1B). For mineralization detection, elevated calcium deposition was observed in the thoracic aorta of BAPN-induced mice compared to controls (Fig. 2E). However, unlike the CaPO₄-AAA model (positive control), von Kossa staining did not detect overt calcification foci in BAPN-induced mice thoracic aorta at either 14 days or 28 days (established AD) (Supplementary Fig. 1C). To overcome the spatial constraints of histology, we employed ¹⁸F-NaF PET/CT imaging in 6 mice, with hydroxyapatite (HAP)-coated aortic arch serving as positive controls. However, no detectable ¹⁸F-NaF uptake was observed in the thoracic aorta of BAPN-induced mice, while the HAP-coated controls clearly showed focal tracer accumulation in the aortic arch (Fig. 2F).

Fig. 2. BAPN-induced TAAD is accompanied by increased aortic TNALP activity and ultrastructural calcium-phosphate deposition.

A Experiment design. Three-week-old mice were treated with 0.3% BAPN in drinking water (0.5 g/kg per day) for 28 days. Normal drink served as negative control. The right panel shows representative images of the aorta from mice after 28 days of BAPN administration, along with corresponding EVG staining. The black arrows indicate the false lumen. Scale bar = 1 mm and 100 μm. B Plasma TNALP levels in control mice and those administered 0.3% BAPN for 14 and 28 days. C Representative Western blot images and corresponding quantitative analysis of TNALP expression in thoracic aortic tissues from mice treated with water (control) or 0.3% BAPN for 14 days. D TNALP activity in the thoracic aorta of control mice and mice treated with 0.3% BAPN for 14 days. E Calcium deposition in the thoracic aorta of control mice and mice treated with 0.3% BAPN for 14 days. F Representative ¹⁸F-NaF PET/CT images of a mouse treated with BAPN for 14 days (top) and a positive control mouse with ex vivo incubation of the aortic arch in HAP for 1 day (bottom). From left to right: CT, ¹⁸F-NaF PET, and merged ¹⁸F-NaF PET/CT images. The dashed lines outline the thoracic aorta, and arrows indicate areas of ¹⁸F-NaF uptake. G Representative TEM images of thoracic aorta from control mice and mice treated with 0.3% BAPN for 14 days. Images are shown from left to right at magnifications of ×8000 (scale bar = 2 μm), ×15,000 (scale bar = 1 μm), and ×50,000 (scale bar = 500 nm). Yellow boxes indicate the regions of interest shown at higher magnification; yellow arrows denote potential calcification foci. H Representative elemental analysis of potential calcification foci using TEM-based energy-dispersive X-ray spectroscopy (EDS). The region indicated by the yellow arrow was identified as a potential calcification focus and imaged at ×50,000 magnification (scale bar = 500 nm). Elemental mapping revealed the spatial distribution of calcium and phosphorus within the selected area. The merged image (bottom right) demonstrates colocalization of calcium and phosphorus, consistent with calcium-phosphate compound deposition. Data presentation: data in (B–E) are presented as mean ± SEM. Each data point represents an individual mouse as a biological replicate with similar results. The sample sizes per group for (B) were control (n = 8), BAPN (n = 8). The sample sizes per group for (C) were control (n = 4), BAPN (n = 4). The sample sizes per group for (D, E) were control (n = 6), BAPN (n = 6). Exact n is provided in the Supplementary Data file. Statistical analysis: statistical analysis of (B–D) was performed using two-sided unpaired Student’s T test. Statistical analysis of (E) was performed using two-sided Mann–Whitney test. Source data are provided as a Supplementary Data file. TAAD thoracic aortic aneurysm and dissection, EVG elastin Van Gieson, BAPN β-aminopropionitrile, TNALP tissue-nonspecific alkaline phosphatase, ¹⁸F-NaF PET/CT Fluorine-18-sodium fluoride positron emission tomography and X-ray computed tomography scanning, HAP hydroxyapatite, TEM transmission electron microscope, EDS energy-dispersive X-ray spectroscopy, VSMC vascular smooth muscle cell, Col collagen.

The absence of detectable calcification via conventional methods (von Kossa staining and ¹⁸F-NaF PET/CT), despite elevated calcium deposition, suggested technical limitations in standard assessment approaches for this model. We therefore used higher-precision transmission electron microscopy (TEM) to evaluate microcalcifications in the mouse thoracic aorta. As shown in Fig. 2G, control VSMCs maintained normal spindle morphology with well-organized collagen fibers and no visible calcium deposition, whereas BAPN-treated mice at 14 days exhibited oval VSMC transformation, disorganized collagen alignment, and scattered dense black spherical foci along collagen fibers. To further characterize the black spherical foci, we performed energy-dispersive X-ray spectroscopy (EDS) analysis. Two candidate foci were identified, with elemental analysis of one focus demonstrating significant enrichment of Ca and P in the foci, indicating the formation of calcium-phosphate complexes (Fig. 2H). Furthermore, cellular alterations, collagen disorganization, and electron-dense foci were consistently observed in the BAPN-induced mice thoracic aortic tissue, suggesting widespread pathological remodeling (Supplementary Fig. 2A). Taken together, these results indicated that aorta mineralization enhanced in BAPN induced TAAD mice.

Calcium-phosphate deposits promote VSMC osteogenic switching and inflammatory remodeling

The origin of aortic calcification remains a long-debated question. VSMC osteogenic differentiation is widely regarded as a source of calcium-phosphate complexes. Runt-related transcription factor 2 (RUNX2), a master regulator of osteogenesis, upregulates TNALP expression through multiple pathways to accelerate calcification^23,27–29. Consistent with the above perspective, we found that BAPN-induced aortic tissues exhibited upregulated RUNX2 and TNALP expression accompanied by reduced contractile markers (SMA, SM22) and elevated inflammatory marker CD68 (Figs. 2C, 3A, B and Supplementary Fig. 3A, B).

Fig. 3. Calcium-phosphate deposits promote VSMC osteogenic switching and inflammatory remodeling.

A, B Representative Western blot images and corresponding quantitative analysis of RUNX2, SMA, and SM22 expression in thoracic aortic tissues from mice treated with water (control) or 0.3% BAPN for 14 days. C Representative SEM images comparing the crystals of calcification in the aortic tissue from an acute aortic dissection AD patient (H-HAP) and industrially synthesized hydroxyapatite (S-HAP). Scale bar = 1 μm and 500 nm. D XRD spectrum of H-HAP (blue) and S-HAP (red). (002), (211), (112), (202), (310), (222), and (213) crystal planes of the standard HAP reference card (JCPDS no. 09-0432). E FT-IR spectrum of H-HAP (blue) and S-HAP (red). F–I VSMCs were treated with 100 μg/mL HAP or S-HAp for 1 day and then washed with PBS to remove HAP and further cultured specified days. F, G Representative Western blot images and corresponding quantitative analysis of TNALP, RUNX2, SMA, and SM22 expression in rat VSMC treated with HAPs for 5 days. H VSMCs were stained for mineralization by Alizarin Red to show calcium deposition after 14 days’ H-HAP or S-HAP treatment. I Quantitative PCR analysis of mRNA expression levels of inflammatory markers (IL-6, IL-1β, MCP-1, and MMP-9) in rat VSMCs after 5 days of treatment with H-HAP or S-HAP. Data presentation: data in (B, G, I) are presented as mean ± SEM. Each data point represents a biological replicate with similar results. The sample sizes per group for (B) were control (n = 6), BAPN (n = 6). The sample sizes per group for (G, I) were control group (n = 6); H-HAP group (n = 6), S-HAP group (n = 6). Exact n is provided in the Supplementary Data file. Statistical analysis: statistical analysis of (B) was performed using two-sided unpaired Student’s T test. Statistical analysis of (G, I) was performed using two-sided Brown–Forsythe and Welch ANOVA tests with Dunnett T3 multiple comparisons. Source data are provided as a Supplementary Data file. BAPN β-aminopropionitrile, TNALP tissue-nonspecific alkaline phosphatase, RUNX2 runt-related transcription factor 2, SMA alpha-smooth muscle actin (α-SMA), SM22 smooth muscle protein 22α, H-HAP human hydroxyapatite, S-HAP synthesized hydroxyapatite, XRD X-ray diffraction, FT-IR, Fourier transform infrared spectroscopy, IL-6 interleukin-6, IL-1β interleukin-1β, MCP-1 monocyte chemoattractant protein-1, MMP-9 matrix metalloproteinase-9.

To further elucidate the potential detrimental effects of aortic calcification on the aorta, calcified components were harvested from the aortic tissue of a patient undergoing aortic replacement surgery for type A AD. These components were subsequently isolated and processed into nanoscale hydroxyapatite (Human-HAP, H-HAP). For comparative analysis, industrially synthesized nano-hydroxyapatite was also employed (Synthesized-HAP, S-HAP). Scanning electron microscopy (SEM) revealed a high degree of morphological similarity between H-HAP and S-HAP, both exhibiting uniform rod-shaped crystalline particles (Fig. 3C). The X-ray diffraction (XRD) spectra of H-HAP and S-HAP displayed characteristic diffraction peaks at 25.8°, 31.8°, 32.9°, 34.0°, 39.8°, 46.7°, and 49.4° (Fig. 3D), corresponding to the (002), (211), (112), (202), (310), (222), and (213) crystal planes of the standard HAP reference card (JCPDS no. 09-0432), respectively. Fourier transform infrared (FT-IR) spectroscopy identified vibration peaks associated with O-H stretching in H-HAP and S-HAP at 3566 and 3432 cm⁻¹, along with asymmetric stretching vibrations of P-O in PO₄³⁻ groups at 564 and 603 cm⁻¹ (Fig. 3E). Both XRD and FT-IR analyses validated that the calcified crystals obtained from patients with aortic dissection consist of pure-phase nanoscale hydroxyapatite. Following treatment with H-HAP and S-HAP, VSMCs exhibited an exacerbated phenotypic transition, marked by TNALP and RUNX2 upregulation, SMA and SM22 downregulation (Fig. 3F, G), and enhanced calcification (Fig. 3H). Meanwhile, inflammatory mediators (IL-6, IL-1β, MCP-1, and MMP9) were significantly elevated (Fig. 3I). These findings suggest that HAP drives VSMCs from a contractile to an osteogenic-like phenotype, accelerates calcium-phosphate deposition, and promotes inflammation, ultimately contributing to aortic wall degradation.

TNALP facilitates cell-independent calcium-phosphate deposition under mineralizing conditions

In our previous experiments, we observed elevated TNALP levels in both plasma and aortic tissues from human subjects and BAPN-induced mice. TNALP catalyzes the hydrolysis of phosphate monoesters, thereby releasing inorganic phosphate, and plays a crucial role in regulating mineral deposition³⁰. Based on these findings, we hypothesize that in addition to VSMC osteogenic differentiation, a parallel passive calcification process driven by TNALP—independent of cellular activity—may also contribute to aortic mineralization. This process could be modulated by various stimuli that upregulate TNALP expression (Fig. 4A). To establish a cell-free in vitro mineralization model, we used type I collagen as the matrix and incubated the system with 10 mM CaCl₂ and 6 mM β-glycerophosphate (β-GP) in the presence of TNALP at 37 °C for 4 h. Alizarin Red staining revealed substantial calcium salt deposition (Fig. 4B). The resulting precipitates were collected and XRD analysis, which demonstrated diffraction peaks (red) that matched those of standard hydroxyapatite (blue), corresponding to the (002), (211), (112), (202), (310), (222), and (213) crystal planes (Fig. 4C, D, and Supplementary Fig. 4A).

Fig. 4. TNALP facilitates cell-independent calcium-phosphate deposition under mineralizing conditions.

A Hypothesis diagram: “Active calcification” refers to the pathological loss of the contractile phenotype in VSMCs under various stimuli, resulting in their osteogenic differentiation, which exacerbates calcium deposition and facilitates the onset of AD. “Passive calcification,” on the other hand, involves the upregulation of TNALP levels (both in plasma and tissues) induced by pathological stimuli, which enhances phosphate release, further promotes calcium deposition, and accelerates the phenotypic transition of VSMCs, ultimately contributing to the occurrence of AD. B Representative images of mineralization assay: type I collagen was coated in 6-well plates, incubated for 30 min, washed with saline, and air-dried. Each well was treated with 2 mL of mineralization solution (10 mM CaCl₂, 6 mM β-GP). Experimental wells received purified TNALP (4 U/mL), while controls received ddH₂O. After 4 h at 37 °C, Alizarin Red staining was performed to evaluate mineralization. C In 20 mL centrifuge tubes, 10 mL of mineralization solution (10 mM CaCl₂, 6 mM β-GP) and ddH₂O/purified TNALP (4 U/mL) were incubated at 37 °C for 8 h. D XRD spectrum of mineralization assay deposits (red), showing characteristic diffraction peaks corresponding to the (002), (211), (112), (202), (310), (222), and (213) crystal planes. The standard hydroxyapatite (HAP) reference pattern (JCPDS No. 09-0432) is shown in blue. E–G Tetramisole reduced plasma TNALP levels, as well as TNALP expression and activity in the aorta of mice. E Experiment design. Three-week-old mice were treated with 0.3% BAPN in drinking water (0.5 g/kg per day) for 14 days. Normal drink served as negative control. Concurrently, tetramisole, a TNALP inhibitor, was administered via intraperitoneal injection at a dose of 16 mg/kg/day. Thoracic aortas were harvest for further analysis. F Plasma TNALP levels of each group. G Representative Western blot images and corresponding quantitative analysis of TNALP expression in thoracic aortic tissues from each group. H In 20 mL tubes, 10 mL of mineralization solution (10 mM CaCl₂, 6 mM β-GP, 4 U/mL TNALP) was prepared. The experimental group received Tetra (1.6 mg/mL), while the control group received ddH₂O. After 8 h at 37 °C, precipitate formation was compared. I XRD spectrum of mineralization assay deposits. The TNALP + Tetra group is shown in red, and the TNALP + ddH₂O group is shown in blue. The standard HAP reference pattern (JCPDS No. 09-0432) is displayed at the bottom of the panel. Data presentation: data in (F, G) are presented as mean ± SEM. Each data point represents an individual mouse as a biological replicate with similar results. The sample sizes per group for (F, G) were water + saline (n = 6); water + Tetra (n = 6); BAPN + saline (n = 6); BAPN + Tetra (n = 6). Exact n is provided in the Supplementary Data file. Statistical analysis: statistical analysis of (F, G) was performed using two-sided Brown–Forsythe and Welch ANOVA tests with Dunnett T3 multiple comparisons. Source data are provided as a Supplementary Data file. TNALP tissue-nonspecific alkaline phosphatase, VSMC vascular smooth muscle cell, β-GP β-glycerophosphate, XRD X-ray diffraction, BAPN β-aminopropionitrile, Tetra tetramisole.

We further investigate whether inhibition of TNALP could attenuate the passive calcification process. We than administered the non-selective TNALP inhibitor tetramisole (Tetra) via intraperitoneal injection^31,32 in BAPN-induced mice for 14 days (Fig. 4E). This intervention resulted in a significant reduction in plasma TNALP levels (water + saline, 287.7 ± 6.0; water + Tetra, 254.9 ± 5.9; BAPN + saline, 381.4 ± 12.9; BAPN + Tetra, 320.8 ± 4.1 U/L), as well as aortic TNALP expression and enzymatic activity (Fig. 4F, G, and Supplementary Fig. 4B–D). Notably, control mice receiving tetramisole also exhibited decreased TNALP levels and activity, suggesting that the inhibitory effect of tetramisole is independent of BAPN exposure. We further performed in mineralization assays with tetramisole intervention and found that inhibition of TNALP significantly reduced calcium salt deposition. XRD analysis revealed that the intensity of diffraction peaks was markedly lower in the tetramisole-treated group compared to the control group (Fig. 4H, I), suggesting that TNALP inhibition effectively reduces both the crystallinity and quantity of HAP formation. Taken together, these findings indicate that TNALP can promote calcium-phosphate complex formation in a cell-independent manner under mineralizing conditions.

Tetramisole-mediated TNALP inhibition reduces aortic calcium-phosphate deposition and VSMC phenotypic switching

We further administered the TNALP inhibitor tetramisole via intraperitoneal injection to BAPN-induced mice for 14 days and observed a significant reduction in thoracic aortic calcium deposition in the BAPN-treated group (BAPN + saline, 0.144 ± 0.004 mmol/g; BAPN + Tetra, 0.096 ± 0.011 mmol/g). In contrast, no statistically significant difference was observed in control mice (water + saline: 0.074 ± 0.006 mmol/g; water + Tetra: 0.065 ± 0.009 mmol/g) (Supplementary Fig. 5A). Furthermore, TEM analysis revealed that, compared to the BAPN + saline group, VSMCs in the BAPN + Tetra group exhibited a more spindle-shaped morphology, with relatively well-preserved and organized elastic fibers and collagen structures. Moreover, calcium-phosphate complex deposition was markedly reduced (Supplementary Fig. 5B). In our previous experiments, we demonstrated that calcium-phosphate complexes could induce osteogenic differentiation (Supplementary Fig. 2F, G). As expected, analysis of mouse aortic tissues revealed a significant attenuation of osteogenic differentiation along with reduced loss of contractile proteins (Supplementary Fig. 5C–F).

Pharmacological inhibition and VSMC-targeted knockdown of TNALP attenuate BAPN-induced TAAD in mice

Finally, we employed a BAPN-induced mouse model of TAAD. Three-week-old male C57BL/6J mice were administered freshly prepared BAPN in their drinking water at a dosage of 1 g/kg per day (approximately 0.5%) for 28 days. Concurrently, a subset of mice received intraperitoneal injections of the TNALP inhibitor tetramisole. Control mice were given normal drinking water (Fig. 5A). There were no significant differences in body weight or water intake among the BAPN-treated groups (Supplementary Table 2). As a result, BAPN treatment induced aortic rupture and premature death in 56.25% (9/16) of the mice. Notably, tetramisole intervention markedly reduced rupture mortality to 20% (3/15) (Fig. 5B, C). Compared with the BAPN + saline group, tetramisole treatment significantly reduced the incidence of TAAD (BAPN + saline, 75% [12/16]; BAPN + Tetra, 33.3% [5/15]) (Fig. 5D, E). Ex vivo measurements revealed significant reduction in the diameters of the ascending aorta, aortic arch, and descending aorta in tetramisole-treated groups compared with that in BAPN + saline group (Fig. 5F–H). EVG staining further revealed that TNALP inhibition markedly attenuated elastic fiber breakage and degradation (Fig. 5I, J). To further address the potential off-target effects of tetramisole and strengthen the TNALP-specific inference, we performed an orthogonal VSMC-targeted knockdown experiment using AAV9-SM22-Sh_ALPL. AAV9-SM22-ShALPL significantly reduced aortic TNALP mRNA expression compared with AAV9-SM22-Sh_Scramble (Supplementary Fig. 6A, B). Consistent with the tetramisole intervention, VSMC-targeted ALPL knockdown improved survival, reduced aortic rupture and TAAD incidence, attenuated aortic dilation, and decreased elastin degradation in BAPN-treated mice (Supplementary Fig. 6C–K). These findings support that suppression of aortic TNALP contributes to the protective phenotype observed in BAPN-induced TAAD. Thus, Inhibition of TNALP attenuates BAPN-induced TAAD in mice.

Fig. 5. Pharmacological inhibition of TNALP attenuate BAPN-induced TAAD in mice.

A Three-week-old mice were treated with 0.5% BAPN in drinking water (1 g/kg per day), accompanied by daily intraperitoneal injections of tetramisole (16 mg/kg/day). After 28 days of treatment, mice were sacrificed for further analysis. B Survival analysis of mice after vehicle/Tetra treatment. *, BAPN + Tetra group vs. BAPN + Vehicle group. C Rupture of TAAD. D Representative ex vivo morphology of the aortas. Scale bar = 2 mm. E Incidence of TAAD. Maximal diameters of F ascending aorta, G aortic arch and H descending aorta. I, J Representative images of elastin Van Gieson staining and statistical analysis on elastin degradation grade. Scale bar = 100/25 μm as shown in the image. Data presentation: data in (B, C, E) are presented as percentage (%). Data in (F–H, J) are presented as mean ± SEM. Each data point represents an individual mouse as a biological replicate with similar results. The sample sizes per group for (A–J) were water + saline (n = 3); water + Tetra (n = 5); BAPN +  saline (n = 16); BAPN + Tetra (n = 15). Exact n is provided in the Supplementary Data file. Statistical analysis: statistical analysis of (B) were performed using two-sided Logrank test. Statistical analysis of (C, E) were performed using two-sided Fisher’s exact test. Statistical analysis of (F–H) were performed using two-sided Brown–Forsythe and Welch ANOVA tests with Dunnett T3 multiple comparisons. Statistical analysis of (J) were performed using Kruskal–Wallis test with Dunn’s multiple comparisons. Source data are provided as a Supplementary Data file. BAPN β-aminopropionitrile, Tetra tetramisole, TAAD thoracic aortic aneurysm and dissection, EVG elastin Van Gieson.

¹⁸F-NaF PET/CT detects focal uptake near intimal tear sites in human AD

In mice, we demonstrated that inhibition of TNALP reduced calcium-phosphate complex deposition and attenuated the incidence of aortic dissection, highlighting the pivotal role of calcific deposition in aortic dissection. To further investigate this association in humans, we enrolled two patients with acute Stanford type B aortic dissection who underwent ¹⁸F-NaF PET/CT imaging immediately after onset to assess the spatial relationship between aortic calcification and the site of intimal rupture.

In the first case (Fig. 6A), a 76-year-old female with a history of intramural hematoma progressed to acute aortic dissection of unknown etiology. CTA revealed a distinct intimal tear (yellow arrow), with prominent macroscopic calcification adjacent to the primary intimal tear site. ¹⁸F-NaF PET/CT imaging revealed notable radiotracer uptake throughout the aortic wall, not only in areas adjacent to macroscopic calcification but also in regions without discernible calcification on CT. Notably, areas of microcalcification exhibited higher ¹⁸F-NaF uptake values, with a maximum standardized uptake value (SUVmax) of 2.78 (yellow arrow), indicating active mineral deposition near the primary intimal tear. The second case (Fig. 6B) involved a 48-year-old male patient who presented with acute type B AD. CTA imaging revealed a clear separation between the true and false lumens, along with an identifiable intimal tear (yellow arrow) and a visible intimal flap (yellow line), but no apparent macroscopic calcification. However, ¹⁸F-NaF PET/CT imaging demonstrated focal radiotracer uptake localized to the region of the intimal flap adjacent to the tear. The uptake intensity was markedly higher than that in surrounding areas at the same axial level, indicating the presence of metabolically active microcalcification at the rupture site. Together, these two cases showed focal ¹⁸F-NaF uptake near intimal tear sites in human AD, suggesting a spatial association between active microcalcification signals and regions of intimal disruption.

Fig. 6. ¹⁸F-NaF PET/CT detects focal uptake near intimal tear sites in human AD.

A, B Representative CTA, CT, ¹⁸F-NaF PET and merge ¹⁸F-NaF PET/CT images from a 76-year-old female patient and 60-year-old male patient with acute aortic dissection. For the CTA image, the true lumen and false lumen of the aorta are labeled. The yellow line indicates the intimal flap, and the yellow arrow marks the intimal tear. For the CT image, the yellow dashed line outlines the aortic wall boundary, and the yellow solid line indicates the intimal flap. For the ¹⁸F-NaF PET and merge image, white arrows indicate areas of macroscopic calcification, while yellow arrows indicate regions of microcalcification. AD aortic dissection, CTA computed tomography angiography, ¹⁸F-NaF PET/CT Fluorine-18-sodium fluoride positron emission tomography and X-ray computed tomography scanning.

Discussion

AC is a hallmark of enhanced mineralization and has been strongly linked to adverse cardiovascular outcomes^33,34. Despite this, its involvement in the pathogenesis of AD has received limited attention. In our previous study, we observed elevated levels of TNALP in AD patients’ plasma, accompanied by increased aortic mineralization. Mendelian randomization analysis further confirmed a genetically mediated contribution of TNALP to AD susceptibility, underscoring a potential pathological axis involving TNALP, AC, and AD⁸. In this study, our findings support the hypothesis that increased TNALP promotes calcium-phosphate microdeposition in the aortic wall, which aggravates VSMC phenotypic switching, inflammatory remodeling, and structural degeneration of the aortic media, thereby increasing susceptibility to AD (Fig. 7).

Fig. 7. Summary diagram.

Elevated TNALP levels and activity—potentially originating from VSMC osteogenic differentiation or pathological metabolic processes in the liver and bone—promote the hydrolysis of organic phosphates and the release of inorganic phosphate. This, in turn, facilitates hydroxyapatite deposition within collagen matrices, further driving VSMC osteogenic transformation, inflammatory responses, and loss of the contractile phenotype, ultimately contributing to the development of aortic dissection.

AD accounts for nearly a half of aortic diseases³⁵. Although AD and aortic aneurysm (AA) share several pathogenic mechanisms—such as VSMC phenotypic switching, elastic fiber degradation, VSMC apoptosis and senescence, endoplasmic reticulum stress, and oxidative stress—they exhibit distinct pathological features and clinical presentations^36,37. Due to the typically slow progression of AA compared to the acute nature of AD, most pharmacological and mechanistic studies have predominantly focused on AA. Previous studies have demonstrated increased AC in both patients with abdominal aortic aneurysm (AAA) and in experimental AAA mouse model, particularly highlighting the presence of active microcalcification (<50 μm)^23,24. Moreover, it is increasingly recognized that microcalcification alter the mechanical properties of the aortic wall, thereby accelerating aneurysm expansion and increasing the risk of rupture^23,24,38. However, evidence on macrocalcification (>50 μm) remains conflicting. While a prospective study in asymptomatic AAA patients suggested a stabilizing role on the vascular wall²⁴, another reported its association with increased overall and cardiovascular mortality in both thoracic and abdominal aneurysms³⁹. Nevertheless, macrocalcification is generally considered to result from the gradual accumulation of microcalcification. Therefore, ¹⁸F-sodium fluoride (¹⁸F-NaF) PET imaging may serve as a valuable tool to assess biologically active microcalcification within the aneurysmal wall, offering potential utility in predicting the risk of AAA-related events²⁴. However, studies on the AC burden and impact in AD remains limited. Our study not only demonstrated a significantly increased AC burden in AD patients’ aorta via CT imaging⁸, but also confirmed the accumulation of microcalcification in aortic tissues from both AD patients and mouse model, consistent with the limited available clinical studies^25,40. For mouse model, AD was induced by BAPN, lysyl oxidase (LOX) inhibitor that disrupts cross-linking of elastin and collagen, thereby promoting aortic dissection formation⁷. Notably, although we observed increasing thoracic aortic calcium deposition in BAPN-induced mice, both von Kossa staining and ¹⁸F-NaF PET/CT yielded negative results. This discrepancy may be explained by the early, sparse, and ultrastructural nature of calcium-phosphate deposition in this model, which may remain below the detection threshold of conventional histological staining and in vivo PET imaging⁴¹. High-resolution imaging by TEM enabled direct visualization of calcium-phosphate deposition and VSMC morphology. In the early stage of BAPN-induced AD in mice, we observed marked alterations in VSMC morphology, disorganized collagen fiber arrangement, and increased deposition of electron-dense granules. To our knowledge, the ultrastructural characteristics of AC in BAPN-induced AD models have not been previously reported. By combining TEM with EDS, we found that representative electron-dense deposits were enriched in calcium and phosphorus, supporting their identity as Ca/P-rich calcium-phosphate complexes⁴¹. However, because crystallographic analyses such as selected-area electron diffraction, XRD, or FTIR were not performed on these sparse ultrastructural deposits, we do not define them as crystalline hydroxyapatite. In contrast, in human AD calcified aortic samples, XRD and FTIR analyses identified hydroxyapatite as the predominant mineral phase, consistent with previous findings in atherosclerosis⁴¹, vascular calcification^42,43, and aortic aneurysms^23,44. Taken together, these findings suggest increased aortic mineral deposition in AD.

VSMCs are key players in aortic calcification. Phenotypic switching of contractile VSMCs into osteo/chondrogenic VSMCs is accompanied by expression of bone-specific proteins that regulate ECM mineralization⁴⁵. Various pathological stimuli or signaling pathways can activate osteo/chondrogenic transcription factors, such as RUNX2, which in turn regulate the expression of downstream effectors, including TNALP, osteocalcin, and osteopontin, thereby promoting calcification deposition^15,46. Consistent with previous findings⁴⁷, our study demonstrated increased expression of TNALP and RUNX2, reduced levels of VSMC contractile markers, and distinct ultrastructural alterations in VSMCs in BAPN-induced mouse. The calcification (hydroxyapatite) has been reported to further promote osteogenic differentiation of VSMC^23,42,43. Therefore, we treated VSMCs with calcified extracts from AD patient aorta and observed enhanced osteogenic differentiation and inflammatory responses. These findings suggest that the “RUNX2-TNALP-calcification” pathway forms a vicious cycle, where any increase in the levels of components within this cycle may exacerbate its progression.

Interestingly, we observed elevated plasma TNALP levels in both AD patients and BAPN-induced mice, a finding that distinguishes our study from other AD-related research. Moreover, our previous Mendelian randomization study identified TNALP as a potential causal factor in the onset of AD⁸. The primary function of TNALP is to hydrolyze pyrophosphate (an inhibitor of calcification) to produce phosphate, which, in the presence of Ca²⁺, promotes the formation of hydroxyapatite¹⁸. TNALP have been reported that directly induce osteogenic differentiation of VSMCs and promote the formation of calcium-phosphate complexes⁴⁸. In clinical trials, treatment of hypophosphatasia patients with recombinant alkaline phosphatase replacement therapy (Asfotase alfa) has been associated with AC side effects^49–51 as well as animal studies⁵². Although TNALP may promote calcium-phosphate deposition through various biological mechanisms, such as osteogenic differentiation of VSMCs, calcification is ultimately a chemically driven process. We therefore hypothesized that TNALP might facilitate calcium-phosphate complex formation in a cell-independent manner by hydrolyzing phosphate monoesters in a mineralizing environment (Fig. 4A). This, concept of “passive calcification” was validated by our in vitro mineralization assay. Similar findings have also been validated in previous acellular experiments^53,54. However, the concentrations of CaCl₂ and β-glycerophosphate used in this assay were adapted from in vitro biomimetic mineralization systems commonly used in dental and hard-tissue research, in which supersaturated mineralizing conditions are applied to induce hydroxyapatite nucleation on collagen-based matrices^55,56. Therefore, this simplified system should be interpreted as a proof-of-principle assay testing the mineralization-promoting capacity of TNALP under permissive conditions, rather than as a direct reproduction of physiological calcium, phosphate-donor, or TNALP concentrations in the aortic wall.

However, the source of elevated plasma TNALP was not determined in this study, and attributing it solely to aortic VSMCs would be premature. TNALP is broadly expressed in multiple organs, including bone, liver, and kidney, suggesting that systemic alterations may contribute to increased circulating TNALP⁵. Consistently, we observed increased TNALP expression in both femur and liver tissues from BAPN-treated mice (Supplementary Fig. 7A) and clinical conditions associated with disturbed mineral metabolism or elevated alkaline phosphatase activity, such as primary hyperparathyroidism and chronic kidney disease, have been linked to vascular calcification and AD-related risk^18,57–61. These observations suggest that extra-aortic sources of TNALP may create a systemic environment permissive for aortic calcium-phosphate deposition, although their precise contribution remains to be determined. Importantly, both pharmacological suppression of TNALP with tetramisole and VSMC-targeted ALPL knockdown using AAV9-SM22-Sh_ALPL reduced the incidence of BAPN-induced TAAD, supporting TNALP as a downstream effector linking mineral dysregulation to AD-associated aortic degeneration. Together with previous evidence that RUNX2 deletion alleviates AngII-induced AAA formation in mice²³, these findings support a contributory role of the RUNX2-TNALP-calcium-phosphate deposition axis in aortic degeneration.

These findings further support the hypothesis that multi-organ dysfunction-induced TNALP elevation may promote pathological biomineralization and subsequently increase AD susceptibility. Although TNALP may originate from multiple sources, we aimed to directly target this terminal effector by administering the TNALP inhibitor tetramisole^31,32 or AAV9-SM22-Sh_ALPL to disrupt the “RUNX2–TNALP–calcification” cycle. We found inhibit TNALP markedly lowered the incidence of AD in BAPN-induced mice. These findings are in strong agreement with previous reports showing that RUNX2 deletion alleviates AngII-induced AAA formation in mice²³. Taken together, TNALP-mediated AC plays a pivotal role in promoting the development of AD.

In our study, we performed ¹⁸F-NaF PET imaging in two patients with AD shortly after symptom onset, one with macroscopic calcification and the other without visible calcification on CT. Both cases showed focal ¹⁸F-NaF uptake near the intimal tear, suggesting a spatial association between active mineralization signals and sites of intimal disruption. However, given the limited number of cases and the post-onset timing of imaging, these observations should be interpreted as proof-of-concept and hypothesis-generating. Increased ¹⁸F-NaF uptake near the tear site may reflect pre-existing active microcalcification, but may also be influenced by acute tissue injury, inflammation, thrombus-associated changes, or early remodeling after AD onset. Larger prospective studies are needed to determine whether ¹⁸F-NaF PET/CT can identify biologically active aortic regions associated with intimal vulnerability or AD-related events.

In summary, our findings support a model in which TNALP-associated calcium-phosphate microdeposition contributes to aortic medial degeneration and AD susceptibility. Detection of active aortic microcalcification may provide a potential approach for assessing aortic wall biological activity, while the therapeutic modulation of TNALP will require further tissue-specific and safety-focused investigation.

Materials and methods

Materials

β-aminopropionitrile monofumarate (BAPN, A3134), Tetramisole (Tetra, T1512) and bovine intestinal alkaline phosphatase (TNALP, p6774) were purchased from Sigma Aldrich (St. Louis, MO, USA). Elastase, porcine pancreas (HY-P2974) was purchased from Med Chem Express (Monmouth Junction, NJ, USA). Type II collagenase was purchased from Yeasen Biotechnology (Shanghai, China). The 5% β-glycerophosphate (G1485) was purchased from Solarbio (Beijing, China). Nano-hydroxyapatite (H106378) was purchased from Aladdin (Shanghai, China). Calcium Assay Kit (C004-2-1) and TNALP Activity Assay Kit (A059-2) were both purchased from Jiancheng Bioengineering Institute (Nanjing, China). The Alizarin Red S staining solution (pH 4.2, C0138 and pH 8.3, C0140) were purchased from Beyotime Biotechnology (Shanghai, China). Antibodies against GAPDH (10494-1-AP) was purchased from ProteinTech (Chicago, IL, USA). Antibody against SM22α (ab10135), SMA (ab5694), RUNX2 (ab114133), CD68 (ab125212), and TNALP (ab108337) were purchased from Abcam (Cambridge, United Kingdom).

Human subjects

Human AD specimens were collected from 5 patients undergoing aortic replacement surgery at the Department of Cardiac Surgery of Chinese PLA General Hospital, Beijing, China. Non-AD aortic tissues were resected from five patients undergoing coronary artery bypass grafting, and did not have AD. Patients with severe conditions, such as chronic kidney disease, collagen vascular disease, and hepatobiliary or bone metabolic disorders, were excluded. After being harvested, aortic tissues were promptly fixed in formalin, embedded in paraffin, and sectioned into 7-μm-thick serial slices for subsequent histological or ¹⁸F-NaF autoradiography.

For human aortic dissection calcification extracts, as previous studies description^42,43, aortic samples were collected from patients undergoing aortic replacement surgery for type A dissection. Calcified or visibly hardened aorta regions were isolated from the vessel wall. The tissue was enzymatically digested overnight at 37 °C using 3 mg/mL collagenase and 1 mg/mL elastase. After digestion, the precipitate was washed three times with PBS to remove residual enzymes. The samples were then passed through a 70 μm filter mesh and rinsed with PBS to separate calcified particles by size. Large particles (>70 μm) retained on the mesh were suspended in PBS and mechanically ground, while the filtrate was ultracentrifuged for 20 min to collect nano-sized calcified particles.

For human ¹⁸F-NaF PET/CT imaging, see the content “¹⁸F-NaF PET/CT imaging in patients and mice.”

The study protocol was approved by the Ethics Committee of the Chinese PLA General Hospital (PLAGH-2025-C-088), and written informed consent was obtained from all patients and their families. All experimental procedures were conducted in strict accordance with relevant regulations and guidelines.

Animals

All animal protocols were approved by the Institutional Animal Care and Use Committee of the Institute of Model Animals of Chinese PLA General Hospital and the Animal Care and Use Committee of Chinese PLA General Hospital (PLAGH 0792). We have complied with all relevant ethical regulations for animal use. SPF mice of C57BL/6J were purchased from SPF Biotechnology (Beijing, China). Mice were randomly allocated into groups and housed in cages (4–6 per cage) with free access to food and water. They were maintained under a 12-h light-dark cycle (lights on at 8 a.m.) at 21–24 °C and 40–70% humidity. Mice for surgeries were anesthetized using isoflurane (1.5–2%). All animals were anesthetized with 1.25% avertin before being sacrificed. Plasma samples were collected via cardiac puncture using heparin sodium (250 U/mL) as an anticoagulant and centrifuged at 3000 rpm for 10 min to obtain plasma. For aortic protein extraction, mice were perfused with ice-cold saline, and the aorta was carefully isolated. For morphological analysis, sequential perfusion was performed with 10% KCl, 10× PBS, and 4% paraformaldehyde. The aorta was then dissected from the root to the iliac bifurcation in all animal models for subsequent examination.

BAPN-induced AD mouse model

Three-week-old C57BL/6J mice were treated with 0.3% BAPN in drinking water (0.5 g/kg per day) for 28 days. The drinking water was refreshed every 3 days. Normal drinking was designed as negative control. The early stage of dissection (defined as no significant macroscopic vascular lesions) was determined at 14 days after BAPN administration, while the endpoint of the model was set at 28 days for statistical analysis of AD phenotypes.

For TNALP inhibitor mouse AD experiment, mice were intraperitoneally injected with tetramisole at a dose of 16 mg/kg/day beginning on the day of BAPN administration and continuing until day 28.

At the sacrificed endpoint, aortas were dissected for ex vivo measurements of the maximal thoracic aorta diameter (ascending aorta, aortic arch and descending aorta). The incidence rate of AD and related mortality were recorded. AD was defined as the presence of a false lumen in the thoracic aorta. For mice with aortic aneurysms (at least 1.5 times extension of aortic diameter), EVG staining was performed to confirm the existence of a false lumen. Overall, any of the following criteria was defined as TAAD: 1. The deceased mice displayed blood clotting in the thoracic cavity. 2. EVG staining revealed aortic false lumen formation.

Elastin Van Gieson staining

Paraffin-embedded sections of the lesion aorta (7 μm thick) were analyzed by elastic Van Gieson staining (Servicebio, Wuhan, China) according to the manufacturer’s protocol for elastin assessment. Elastin degradation graded as grade 1, <25% degradation; grade 2, 26–50% degradation; grade 3, 51–75% degradation; grade 4, >75% degradation.

Calcium assay (aorta calcification)

Briefly, the thoracic aortas from BAPN or BAPN-Tetra models were collected for further analysis. The calcium was determined using calcium colorimetric assay kit from Jiancheng Bioengineering Institute (Nanjing, China) according to the manufacturer’s protocol and normalized to total protein measured by the BCA assay.

Alizarin Red or von Kossa staining for characterization of calcified nodules

Alizarin Red staining was used for detection of both VSMCs and collagen mineralization deposition. Specifically, alizarin red (PH = 4.2) was used to stain calcified nodules in VSMCs, while Alizarin Red (PH = 8.3) was employed to stain calcified nodules in the collagen mineralization deposition, due to the different sources of calcification. Briefly, VSMCs or collagen in the culture dish were gently washed with saline, stained with 1 ml of Alizarin Red solution for 10 min, and then the staining solution was removed. After washed three times with saline, images were captured for record.

For von Kossa staining, the thoracic aortas in human, BAPN-model, CaPO₄-model (positive control) were collected for paraffin-embedded sections. Sections of aorta were cut into 7 μm in thickness and hydration treatment. The sections were treated with a 5% silver nitrate (AgNO₃) solution and exposed to ultraviolet light for 15–30 min until optimal color development was achieved. After removing the AgNO₃ solution, the sections were rinsed with 5% sodium thiosulfate (Na₂S₂O₃). Nuclei were then counterstained using nuclear fast red. Calcified nodules appeared as brown to black deposits under microscopic examination.

Masson staining

Paraffin-embedded sections of the aorta were cut into 7 μm in thickness and hydration treatment. The section was stained in accordance with the manufacturer’s protocol using Trichrome stain kit (G1006, Servicebio, Wuhan, China).

Alkaline phosphatase activity

Briefly, the thoracic aortas from BAPN or BAPN-Tetra models were collected for further analysis. TNALP activity was measured using a fluorescence detection kit (A059-2) from Jiancheng Bioengineering Institute (Nanjing, China) in tissue extracts according to the manufacturer’s protocol. The standard curve was plotted using p-nitrophenol, and TNALP activity of each sample was normalized to the total protein concentration.

Serum alkaline phosphatase analysis

Blood samples were collected from the right ventricle of mice and centrifuged to obtain plasma. The supernatant was carefully aspirated, and serum alkaline phosphatase levels were quantified using an automated biochemical analyzer provided by Servicebio (Wuhan, China).

Immunohistochemistry (IHC) and immunofluorescence (IF) staining

Paraffin-embedded sections of the aorta were cut into 7 μm in thickness and hydration treatment. For immunohistochemical staining, endogenous peroxidase activity was blocked by incubating the sections in 3% hydrogen peroxide solution at room temperature for 25 min in the dark. The sections were blocked with 3% BSA and incubated with primary antibodies against TNALP (1:100, Abcam), RUNX2 (1:50, Abcam) at 4 °C overnight, followed by incubation with a horseradish peroxidase-labeled anti-rabbit IgG secondary antibody before staining with DAB (3,3′-diaminobenzidine) Kit (Servicebio, Wuhan, China). Nuclei were counterstained with hematoxylin. For immunofluorescence, primary antibodies against TNALP (1:100, Abcam), RUNX2 (1:50, Abcam), CD68 (1:100, Abcam), SM22α (1:300, Abcam), SMA (1:300, Abcam) at 4 °C overnight, followed by incubation with a CY3-conjugated goat anti-rabbit IgG secondary antibody. Nuclei were counterstained with DAPI.

Rat vascular smooth muscle cell culture

Primary rat VSMCs were isolated from the thoracic aortas of 150 to 180 g male Sprague-Dawley rats by collagenase digestion as previously described⁶². VSMCs were cultured in low glucose DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 μg/ml streptomycin. Primary VSMCs at passages 3–6 were used for all experiments. Cells were maintained at 37 °C in a humidified atmosphere with 5% CO₂.

For VSMC trans-differentiation and inflammation assays, VSMCs were treated with either H-HAP (100 μg/mL) or S-HAP (100 μg/mL) for 1 days, and then washed with PBS to remove HAP and further cultured for 3 days. Western blotting was performed to assess the expression levels of TNALP, RUNX2, SMA, and SM22, while qPCR was conducted to evaluate the mRNA expression of inflammatory markers, including IL-6, IL-1β, MCP-1, and MMP-9.

For Alizarin Red staining, VSMCs were treated with either H-HAP (100 μg/mL) or S-HAP (100 μg/mL) for 1 days, and then washed with PBS to remove HAP and further cultured for 14 days.

Hydroxyapatite application in mice thoracic aorta

Nano-hydroxyapatite (from Aladdin, Shanghai, China) was suspended in 30% Pluronic F-127 gel at a concentration of 0.25 M and applied to the ascending aorta of mice. The incubation procedure was performed as previously described⁶³. Briefly, after anesthetizing the mice, a midline thoracotomy was performed to expose the ascending aorta. Thirty microliters of the hydroxyapatite gel were applied topically to the ascending aorta. Then, the mice were sutured and recovered for further application.

Calcium phosphate-induced AAA model

As previously described⁶⁴, 8-week-old C57BL/6J mice were used for the model. The infrarenal abdominal aorta was wrapped with gauze soaked in 0.5 M CaCl₂ for 10 min, followed by another gauze soaked in PBS for 5 min to induce in situ formation of CaPO4 crystals on the aortic adventitia. After 14 days, the aorta was harvested and preserved in paraffin-embedded sections for further analysis.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Thoracic aortas were grinded by Grinding instrument (Jing N·9548, HODER, Beijing, China) with 65 Hz and 30 s, while VSMCs were washed by saline 3 times. Total RNA was extracted using TRIzol reagent (Life Technologies, CA, USA) according to the manufacturer’s instructions. Equal amounts (1 μg) of RNA were reverse-transcribed to cDNA using NovoScrip® Plus AII-in-one 1st Strand cDNA Synthesis SuperMix (E047-01B, NOVOPROTEIN, Shanghai, China). SYBR Green 2×PCR mix (TransGen Biotech, Beijing, China) was used according to the manufacturer’s instructions. The amplification reactions were programmed at 95 °C for 2 min followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. The data were analyzed using the ΔΔCT method with QuantStudio Design & Analysis Software (Thermo Fisher Scientific, Waltham, MA, USA). Mouse samples were normalized to GAPDH, and Rat samples were normalized to β-actin. The primer sequences for target genes are listed in Supplementary Table 3.

Western blot

Mouse tissues or cells were lysed in RIPA buffer (P0013B, Beyotime, Beijing, China) supplemented with protein phosphatase inhibitors (1:100). Protein concentrations were quantified using a BCA Protein Assay Kit (Thermo Scientific). Equal amounts of total proteins were resolved on 8% or 12% SDS-PAGE gels transferred onto nitrocellulose filter membranes (Pall, Port Washington, NY, USA). The membranes were blocked with 5% BSA and then incubated with the primary antibody and the IRDye 700/800DX-conjugated secondary antibody (Rockland Inc., Gilbertsville, PA, USA). The immunofluorescent signal was obtained using an Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA).

Adeno-associated virus 9 (AAV9) production and infection

A plasmid encoding promoter SM22α, EGFP and shRNA targeting mouse ALPL was engineered by cloning the following shRNA into the AAV9 vector: Sh_ALPL (sense)

5′-GGAAATACATGTACCCGAAGACTCGAGTCTTCGGGTACATGTATTTCC-3′;

Packaging and producing work were performed by GENE company (Shanghai, China). For in vivo infection experiments, every mouse was intravenously administered AAV9-Sh_Scramble and AAV9-Sh_ALPL with 2 × 10¹¹ v.g. titer. Infection efficiency was analyzed in aorta samples by qPCR.

¹⁸F-NaF autoradiography

Paraffin-embedded tissue sections were deparaffinized in xylene (2–3 min) and rinsed with ethanol to remove residual xylene. After drying, sections were incubated with 1 mL of ¹⁸F-NaF solution (10 μCi/mL in <10% ethanol-water) for 1.5 h at room temperature. Sections were then washed sequentially with 50, 40, and 20% ethanol to remove unbound tracer and nonspecific binding. After drying, sections were wrapped in plastic film and exposed to a phosphor screen in the dark for 30–60 min. Radiolabelled sections were scanned using a laser confocal optical scanner, and images were analysed using OptiQuant™ software.

¹⁸F-NaF PET/CT imaging in patients and mice

For patients ¹⁸F-NaF PET/CT imaging, Patients received an intravenous injection of ¹⁸F-NaF (3.70–5.55 MBq/kg) via the arm and rested in a quiet room for 45–60 min before imaging. The bladder was emptied prior to the scan. Images were acquired using a dedicated PET/CT scanner (United Imaging uEXPLORER). A low-dose CT scan (for attenuation correction and anatomical reference) was performed from the head to the feet (or at least including the knees), with patients in a supine position and arms raised above the head. CT localization parameters: 80 kV, 20 mA; CT imaging parameters: 120 kV, 80–100 mAs, slice thickness 3.0 mm, and spacing 1.5 mm. PET imaging was conducted in 3D mode with a 600 mm field of view, 5 min per bed position, slice thickness 2.886 mm, matrix 192 × 192, 3 iterations, and 20 subsets. PET/CT images were analysed using Carestream software version 12.1.5.7014 (Carestream Healthcare, Rochester, NY, USA). On co-registered PET and CT images, ¹⁸F-NaF uptake in the aortic wall structures (true lumen side, intimal flap, and false lumen side) was observed and compared, with quantification performed using standardized uptake values (SUV).

For mice ¹⁸F-NaF PET/CT imaging, according to previous description^23,65, mice were intravenously with 0.1 mL of ¹⁸F-NaF (~10 MBq) in saline. After 10 min, Micro PET-CT imaging (Pingsheng Medical Technology, Kunshan) was performed under isoflurane anaesthesia (3% induction at 1.0 L/min, maintained at 0.8 L/min). Mice were positioned supine for CT acquisition, followed by PET scanning (axial length: 127 mm). Post-scan, mice were recovered in a lead-shielded chamber. Image data were processed using Recon/Avators-10 software, and ¹⁸F-NaF uptake was analysed.

Mineralization assay

Mineralization Assay 1

Type I collagen was coated in 6-well plates, incubated for 30 min, washed with saline, and air-dried. Each well was treated with 2 mL of mineralization solution (10 mM CaCl₂, 6 mM β-glycerophosphate). Experimental wells received purified TNALP (4 U/mL), while controls received ddH₂O. After 4 h at 37 °C, Alizarin Red staining was performed to evaluate mineralization.

Mineralization Assay 2

In 20 mL centrifuge tubes, 10 mL of mineralization solution (10 mM CaCl₂, 6 mM β-glycerophosphate) and ddH₂O/purified TNALP (4 U/mL) were incubated at 37 °C for 8 h. Precipitate formation was observed, followed by centrifugation at 5000 rpm for 20 min. The pellet was washed with ddH₂O and 50% ethanol, dried at 80 °C for 2 h, and ground into powder for further analysis.

Mineralization Assay 3

In 20 mL tubes, 10 mL of mineralization solution (10 mM CaCl₂, 6 mM β-GP, 4 U/mL TNALP) was prepared. The experimental group received Tetra (1.6 mg/mL), while the control group received ddH2O. After 8 h at 37 °C, precipitate formation was compared. Precipitates were collected for further analysis.

Transmission electron microscopy (TEM) analysis

Mouse aortic tissue samples were dehydrated using a graded acetone series (50, 70, 90, and 100%), with three 10-min treatments at each concentration. Infiltration was performed sequentially with acetone and embedding medium at ratios of 1:1, 1:2, and 1:3, each for 1 h, followed by embedding in EPON812 resin and polymerization at 60 °C for 48 h. Ultrathin sections (70 nm) were cut using a Leica UC6 ultramicrotome. Sections were stained with uranyl acetate and lead citrate, then observed and imaged using a transmission electron microscope (JEM-1400). Elemental composition of target regions was further analyzed using energy dispersive spectroscopy (EDS) on a JEM-2100F.

Scanning electron microscopy (SEM) analysis

The microstructures of human aortic dissection calcification (H-HAP) samples were observed using scanning electron microscopy (HITACHI S4800; Nova Nano 450) and compared with industrially synthesized nano-hydroxyapatite (S-HAP). Samples were mounted on copper stubs using conductive adhesive and sputter-coated with gold (10 mA, 3 min) to enhance conductivity. Scanning parameters were set as follows: acceleration voltage of 10 kV, beam current of 500 pA, working distance of 8 mm, and scan speed of 1 frame per second. A secondary electron detector was used to ensure high-resolution imaging and detailed surface morphology.

X-ray diffraction (XRD) analysis

The crystalline structures of precipitates from in vitro mineralization experiments, human aortic calcification samples, and industrially synthesized hydroxyapatite were analyzed using a D8 ADVANCE X-ray diffractometer (XRD). A Cu target was used as the X-ray source, with an acceleration voltage of 40 kV and a current of 40 mA. The scanning parameters were set to a speed of 0.1°/s, covering an angular range of 10°–90°.

Statistics and reproducibility

Statistical analyses were performed using GraphPad Prism 8.0 software (San Diego, CA, USA). Data are expressed as mean ± SEM. The details of the statistical analysis used in each experiment are presented in the corresponding figure legends. For statistical comparisons, we first evaluated whether the data were normally distributed using the Shapiro–Wilk normality test. For two groups, we used unpaired Student’s t test for normally distributed data. If the data was skewed distribution, we used the Mann–Whitney test. For more than two groups, we used one-way ANOVA followed by post hoc analysis for comparisons among normally distributed data. Specifically, Dunnett’s multiple comparison test was used to compare each group with a control group, while Tukey’s multiple comparisons test was used to compare each group with every other group. If the data equal variances were not assumed, Brown–Forsythe and Welch ANOVA test was used with Dunnett T3 multiple comparisons test (compare each group with a control group) or Games-Howell multiple comparisons test (compare each group with every other group). Nonparametric tests (Kruskal–Wallis test) were used when the data were not normally distributed. In addition, Fisher’s test was used for contingency data, and log-rank test was used the Kaplan–Meier curve.

In all analyses, P < 0.05 was regarded as statistically significant.

Author contributions

W.G., H.P.Z., R.M.W., Y.H., and L.C. conceptualized and designed the study. W.G., L.C., and T.F.M. supervised the study. T.F.M., L.C., H.C., and X.B. performed the experiments and analyzed the data. L.C. and T.F.M. wrote the manuscript with input from all authors. All authors edited the manuscript and approved the final manuscript.

Peer review

Peer review information

Communications Biology thanks Jianyun Yan and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary handling editors: Ophelia Bu and Nilanjan Banerjee. A peer review file is available.

Funding

This research was supported by funding from the National Key R&D Program of China (2024YFC2419000 to W.G.) and the National Natural Science Foundation of China (82170494 to W.G.). Science and Technology Project of Tianjin Municipal Health Commission (TJWJ2023QN116 to L.C.). Yanzhen Talents Scheme supported by Zhengzhou Joint Logistics Support Center of PLA (L.C.).

Data availability

The source data underlying the graphs and charts in the main and Supplementary Figs. are provided in Supplementary Data. Original blot images are provided in the Supplementary Information. All other data supporting the findings of this study are available within the article and its Supplementary Information.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s42003-026-10871-z.