Male C57BL/6 (H-2b) and BALB/c (H-2d) mice (8–10 weeks old) were purchased from Charles River (Beijing, China). Pcsk9^-/- mice (C57BL/6-Pcsk9^eml1Smoc, stock no. NM-KO-200408) and hPCSK9 mice (C57BL/6-Pcsk9^{em2/(hPCSK9)/Smoc}, stock no. NM-HU-00075) were generated by Shanghai Model Organisms (Shanghai, China). All animal experiments were approved by the Animal Care and Use Committee of Huazhong University of Science and Technology. All procedures were performed according to the NIH Guide for the Care and Use of Laboratory Animals.

Abdominal aortic transplantation was performed as previously described (16). For the allograft group, abdominal aortas of male BALB/c (H-2d) mice were orthotopically transplanted into fully major histocompatibility complex (MHC)-mismatched male C57BL/6 (H-2b) recipients, B6 background hPCSK9 or Pcsk9^-/- recipients. For the isograft group, abdominal aortas of C57BL/6 (H-2b) mice were orthotopically transplanted into MHC-matched C57BL/6 (H-2b) recipients. To investigate the role of PCSK9 in GVD, Isograft group (WT B6 to WT B6), WT Allograft group (BALB/c to WT B6), and Pcsk9^-/- Allograft group (BALB/c to Pcsk9^-/- B6) were constructed. We also performed abdominal aortic transplantation on humanized PCSK9 mice to explore the role of PCSK9 inhibitors (Evolocumab) in GVD, and the following groups were included: Isograft group (hPCSK9 B6 to hPCSK9 B6 treatment with vehicle), IgG Group (BALB/c to hPCSK9 B6 treatment with isotype IgG), and Evolocumab group (BALB/c to hPCSK9 B6 treatment with Evolocumab).

Serum samples were harvested from recipients at 8 weeks after surgery. The serum level of PCSK9 was measured by an ELISA kit according to the manufacturer’s protocol. The absorbance was measured at 450 nm. The concentration of PCSK9 was calculated based on the optical density value and the standard curve.

qRT-PCR was performed as previously described (17). Total RNA was isolated from graft vessels using Total RNA Extraction Kit (Solarbio, R1200, China) according to the manufacturer’s protocol. cDNA was synthesized by the cDNA synthesis kit (Abclonal, RK20400, China), according to the manufacturer’s instructions. Q-PCR were then performed using RealSYBR Mixture kit (CWBIO, CW0760M, China) according to the protocol. Gapdh was used as an internal control. The primer sequences are shown in Table Sl . Detailed information of critical commercial kits are showed in Supplementary Table S2 .

Graft vessels, liver, small intestine, and kidney were harvested at 2 or 8 weeks (time points have been explained in figure legends of each figure) after transplantation. And these samples were used for difference experiments, including Hematoxylin & eosin (H&E) staining, Masson’s staining, EVG staining, immunohistochemical staining, and immunofluorescent staining as previously described (18). ImageJ software was used to determine the intima-media (I/M) ratio, areas of intimal hyperplasia, areas of Masson’s staining and immunohistochemical quantitative analysis. During the quantitative analysis of collagen deposition and immunohistochemical, entire area has been measured.

Western blot was performed as previously described (17). Total protein was extracted from cells or tissue samples using RIPA lysing buffer. The protein concentration of each sample was measured using the bicinchoninic acid protein assay kit. Western blot was performed according to a protocol of our laboratory. The membranes were separately incubated with the following primary antibodies. The antibodies used for Western blot are shown in Table S2 .

VSMCs and BMDMs were cultured as previously described (19). Descending thoracic aortas were collected from mice at 8–10 weeks under sterile conditions. The aortas were then minced, and treated with 1% type I, II, IV collagenase (Worthington Biochemical Corporation) and 1% dispase for 1 h at 37°C. The cells were then harvested and resuspended in Dulbecco’s modified Eagle’s medium (DMEM) containing 20% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Cells at passage 4–8 were used for the in vitro experiments. The purity of VSMCs was identified. Born marrow were extracted from the femurs and tibias of age-matched WT and Pcsk9^-/- mice and then flushed with cold DMEM. After lysis of red blood cells, the BMDMs were harvested and resuspended in DMEM medium containing 10% FBS, 1% penicillin/streptomycin, and 20 ng/mL macrophage colony-stimulating factor (M-CSF; Peprotech, 315-02-10UG). The culture medium was replaced every other day with fresh medium containing 20 ng/mL M-CSF. After seven days of differentiation, the cells were counted and plated for further experiments.

A transwell assay was performed to detect the migration capacity of VSMCs and BMDMs. Cells (three replicates per group) were added to the upper chamber (filled with serum-free medium) of a 24-well transwell plate (Corning 3422) at a density of 2 × 10⁴ cells/well. Serum samples were collected from allograft or isograft mice at 2 weeks after transplantation. A volume of 600 μL of culture medium supplemented with 10% serum was used for cell chemotaxis. After incubation for 48 h, the cells were fixed with 4% paraformaldehyde. Then, the cells on the upper side were wiped off. The invaded cells were stained with crystal violet (0.1%) solution for 10 min. Subsequently, the cells were visualized under a fluorescence microscope. Five random fields per slide were photographed. All experiments were performed at least three times, with each performed in duplicate.

The proliferation rates of BMDMs and VSMCs were examined using a BeyoClick™ EdU Cell Proliferation Kit with Alexa Fluor 488 (Beyotime, C0071S), according to the manufacturer’s protocol. Briefly, serum was collected from allograft or isograft mice at 2 weeks after transplantation. Cells at a density of 5 × 10⁴ were plated in 24-well plates and incubated with 10% serum for 48 h. Then, the cells were incubated with 10 μM Edu for 6 h, fixed in 4% paraformaldehyde, and stained with Hoechst 33342. Finally, EdU-positive cells were counted by fluorescence microscopy.

Data are expressed as the mean ± SD (standard deviation). Unpaired t-test was used to compare two independent groups. One way ANOVA followed by Turkey’s post-hoc test or two-way ANOVA was used for comparisons among multiple groups. All statistical analyses were performed using SPSS 19.0 software. Plots were generated using GraphPad Prism 7.0a software (GraphPad Software, Inc., San Diego, CA) for Mac OS X. The significance level was determined at p < 0.05.

To explore whether PCSK9 is related to GVD, we measured the expression of PCSK9 in mice undergoing abdominal aorta transplantation. The abdominal aortas of BALB/c or C57BL/6 (B6) mice were orthotopically transplanted into B6 recipients. Serum samples and graft vessels were harvested at 8 weeks after surgery when allograft vasculopathy occurred. The allograft group showed a significantly higher level of PCSK9 in the serum compared to the isograft group (252.41 ± 68.32 ng/mL vs. 657.19 ± 198.34 ng/mL; p < 0.05). Similarly, in the graft vessels, the expression of PCSK9 in the allograft group was significantly greater than that of the isograft group ( Figure 1A ). PCSK9 is mainly secreted by the liver, but it is also expressed in extrahepatic tissues such as the intestine and kidney. Therefore, we further assessed the expression of PCSK9 in the liver, small intestine, and kidney of mice. The allograft group showed higher PCSK9 expression in the liver compared to the isograft group ( Figures 1B–D ), while no significant difference was observed in PCSK9 expression in the small intestine or kidney between the two groups ( Figure S1 ; Figures 1E, F ). The immunohistochemical staining results were consistent with those of western blot analysis ( Figure 1G ). These findings indicate that PCSK9 is involved in the pathogenesis of GVD.

To investigate the role of PCSK9 in GVD, we generated Pcsk9^-/- mice. The abdominal aortas of BALB/c mice were orthotopically transplanted into Pcsk9^-/- mice. Graft vessels were harvested at 8 weeks after surgery ( Figure 2A ). The histological analysis revealed that the Pcsk9^-/- group had significantly less allograft vasculopathy ( Figure 2B , I/M ratio: 4.34 ± 0.52 vs. 1.51 ± 0.50; p < 0.05) and a significantly smaller neointimal area ( Figure 2C , 15.88 ± 2.09 × 10⁴ μm² vs. 8.29 ± 1.35 × 10⁴ μm²; p < 0.05) compared with the WT group. Masson’s trichrome staining showed that collagen deposition in the allograft group was dramatically increased relative to the isograft group, whereas Pcsk9 knockout significantly reduced the collagen deposition in the allograft mice ( Figure 2D , 24.11 ± 5.72 × 10⁴ μm² vs. 9.98 ± 1.82 × 10⁴ μm²; p < 0.05). Altogether, these data suggest that Pcsk9 depletion attenuates allograft vasculopathy in a mouse model of GVD.

T cells and macrophages play a dominant role in GVD (20, 21). The infiltration of T cells and macrophages into the graft was detected by immunohistochemistry. Pcsk9 knockout significantly reduced the infiltration of F4/80⁺ macrophages into the graft vessels ( Figure 3A ). However, there was no significant difference in CD3⁺ T cell infiltration between the Pcsk9^-/- and WT groups ( Figure 3B ). We further analyzed the mRNA expression levels of proinflammatory cytokines, including Il1b ( Figure 3C ), Il6 ( Figure 3D ), Il18 ( Figure 3E ), Ifng ( Figure 3F ), Tnf ( Figure 3G ), Tgfb ( Figure 3H ), Ccl2, ( Figure 3I ), and Vcam1 ( Figure 3J ), in graft vessels. Il1b, Il6, Il18, Ifng, Tgfb, Ccl2, and Vcam1 were downregulated in the Pcsk9^-/- group compared with the WT group. These findings indicate that Pcsk9 knockout reduced the infiltration of macrophage and inhibited the expression of proinflammatory factors in graft vessels.

The migration and proliferation of VSMCs are key factors in allograft vasculopathy (22, 23). In addition, neointimal smooth muscle cells are mainly derived from the host (24). To explore the effect of PCSK9 on VSMCs migration and proliferation, we extracted VSMCs from WT and Pcsk9^-/- mice. VSMCs were stimulated with the serum collected from recipient mice for 48 h. Transwell assay was performed to determine the migration capacity of VSMCs ( Figure 4A ). Pcsk9 knockout significantly inhibited the migration of VSMCs ( Figures 4B, C , 238.80 ± 21.94 vs. 148.60 ± 27.19; p < 0.05). Moreover, EdU staining showed that Pcsk9 depletion inhibited the proliferation of VSMCs ( Figures 4D, E , 185.40 ± 13.24 vs. 110.60 ± 17.89; p < 0.05). We also found that PCNA and MMP9 were downregulated in Pcsk9-knockout VSMCs ( Figures 4F, G ). However, Pcsk9 knockout did not affect the migration or proliferation of macrophages compared with the control group ( Figure S2 ). These results demonstrate that Pcsk9 knockout inhibits the migration and proliferation of VSMCs stimulated with the serum collected from post-transplant mice.

To further explore whether PCSK9 is involved in the signaling pathways related to proliferation and inflammation, we isolated VSMCs from WT and Pcsk9^-/- mice and treated them with control or allograft serum for 24 h. Western blot analysis ( Figures 5A, B ) suggested that Pcsk9 depletion did not affect the MAPK-MEK1/2 pathway and the TGF-β/Smad3 pathway. However, Pcsk9 knockout inhibited the NLRP3 inflammasome pathway and phenotypic modulation of VSMCs ( Figures 5C–E ). Furthermore, graft vessels were harvested at 8 weeks after transplantation and analyzed by western blot and immunohistochemistry examination. The results showed that the expression levels of NLRP3 and IL-1β in the Pcsk9^-/- group were lower than those in the WT group ( Figures 5F–H ). Collectively, these data imply that Pcsk9 knockout inhibited the activation of NLRP3 inflammasomes in VSMCs.

Evolocumab, a neutralizing antibody to human PCSK9, has been tested in clinical trials for familial hypercholesterolemia or hyperlipidemia that cannot be controlled by statins (25). To investigate whether Evolocumab would be an effective treatment for GVD, we performed allogeneic abdominal aorta transplantation surgery, during which the abdominal aortas of BALB/c mice were orthotopically transplanted into B6 background humanized PCSK9 mice (hPCSK9) recipients. After transplantation, the recipients were subcutaneously injected with Evolocumab (10 mg/kg) or IgG (10 mg/kg) every 2 weeks for 8 weeks ( Figure 6A ). Western blot ( Figures S3A, B ) and immunofluorescence analyses ( Figures 6B, C ) showed that LDLR was significantly upregulated in the Evolocumab-treated group at 8 weeks after transplantation. Graft vessels were collected for histological staining ( Figure 6D ), and the histological analyses showed that Evolocumab-treated mice had significantly reduced vascular stenosis ( Figure 6E , I/M ratio: 5.16 ± 1.16 vs. 2.86 ± 1.43; p < 0.05) and a significantly smaller area of intimal hyperplasia ( Figure 6F , 17.10 ± 1.28 × 10⁴ μm² vs. 8.15 ± 1.46 × 10⁴ μm²; p < 0.05) compared with the IgG-treated group. Masson’s trichrome staining showed that Evolocumab treatment significantly decreased collagen deposition in the hPCSK9 mice ( Figure 6G , 19.16 ± 1.74 × 10⁴ μm² vs. 9.86 ± 3.39 × 10⁴ μm²; p < 0.05). Taken together, Evolocumab inhibited the progression of GVD in hPCSK9 mice.