PIAS3 gene (NM_001165949) is located on mouse chromosome 3 with fourteen exons (15). PIAS3 deficient (PIAS3^−/−) mice on C57BL/6 genetic background were created in Cyagen Biosciences (Suzhou) Inc (Taicang, Jiangsu, China) using the CRISPR/CAS9 technique. Briefly, the exons of 2–9 were selected as target sites, and two gRNAs with Cas9 mRNA were microinjected into zygotes for PIAS3 deficient mouse generation (Figure 1A). F0 founders were identified via PCR genotyping followed by DNA sequencing. Homozygotes (PIAS3^−/−) and wild type (PIAS3^+/+) littermates were screened and used for all experiments. The use and care of animals as well as all experimental procedures were reviewed and approved by the Laboratory Animal Administration Committee of Xi'an Jiaotong University (No. 2019–1178). All information on animals and related reagents were detailed in Supplementary Table S1.

For PCR screening, tail tip samples were collected from <3 weeks old mice and genomic DNA was extracted by proteinase K digestion. Genotyping PCR primers were AAACAAGACTAAAGGAGTATGGGC (sense 1), TAGAGGAAGGGGAAGGGAACTAAG (antisense 1) and CTCAGACACTCGGAAACTCATC (antisense 2). For quantitative reverse transcription PCR (qRT-PCR) and Western blotting analyses, total RNA and proteins were extracted from aortas. The PIAS3 primers for qRT-PCR analysis were sense: ACCAAGAATGGAGCTGAGCC (sense) and TCTGGATTCCGGATCCCCTT (antisense). Antibodies against PIAS3 and β-actin were obtained from Cell Signaling Technology (Danvers, MA, United States) and TransGen Biotech Co., Ltd (Beijing, China), respectively.

Experimental AAAs were induced in PIAS3^−/− and PIAS3^+/+ mice by transient luminal infusion of porcine pancreatic elastase (PPE, 1.5 units/mL in phosphate-buffered saline) in controlled infrarenal aortic segment as previously reported (16, 17). Briefly, 9–12 weeks old male mice were anesthetized by 2% isoflurane inhalation, and infrarenal aorta was exposed via a laparotomy. Infrarenal aorta was then infused with PPE solution via an aortotomy for 5 min under constant pressure as previously described (18). After the surgery, all mice were maintained in individual cages with free access to chow diet and water.

Prior to PPE infusion, infrarenal aorta was photographed with a digital camera. The diameter was measured using Images Plus3.0 ML (Motic Electric Group Co., Ltd, Xiamen, Fujian, China) and served the baseline level. On day 14 after PPE infusion, infrarenal aortic diameter for each mouse was measured by the same procedure prior to sacrifice and aorta was then harvested. A mouse with a more than 50% aortic dilation over the baseline was considered aneurysmal (18).

Mice were euthanized by carbon dioxide inhalation. PPE-infused aortic segment was collected, embedded in optimal cutting temperature media, and sectioned (6 μm). Frozen sections were stained with hematoxylin and eosin (H & E) and Elastic van Gieson (EVG) stains, respectively, for the assessment of general morphology and elastin integrity. Aortic media elastin degradation was graded as I (mild) to IV (severe) on EVG-stained sections as previously reported (16–20). To evaluate SMC depletion, acetone-fixed frozen sections were stained with a goat anti-SMC α-actin antibody followed a standard biotin-streptavidin peroxidase procedure. Biotinylated donkey anti-goat IgG antibody and streptavidin-peroxidase conjugate were obtained from the Jackson ImmunoResearch Laboratories Inc., West Glove, PA, United States). AEC substrate kit for color development was purchased from Vector Laboratories Inc., Burlingame, CA, United States. Aortic medial SMC depletion was graded on a scale of I (mild) to IV (severe) as reported previously (20).

Acetone-fixed aortic frozen sections were stained with monoclonal antibodies against CD68 (macrophages), CD4⁺ T cells, CD8⁺ T cells and B cells (B220) (18). Sections were sequentially incubated with biotinylated goat anti-rat antibody (Vector Laboratories Inc), PBS, streptavidin-peroxidase conjugate (Jackson ImmunoResearch Laboratories Inc), PBS, peroxidase substrate AEC (Vector Laboratories Inc), counterstained with hematoxylin, mounted and cover-slipped. Aortic mural macrophage accumulation was graded on a scale of I (mild) to IV (severe) (20). CD4⁺ T cells, CD8⁺ T cells and B cells were quantitated as positively stained cells per aortic cross section (ACS) (20).

Matrix metalloproteinases (MMPs), particularly MMP2 and MMP9, contribute to the pathogenesis of AAAs. MMP2 and MMP9 were assessed by immunostaining using goat anti-mouse polyclonal antibodies against MMP2 (AF1488) and MMP9 (AF909) (R & D Systems, Minneapolis, MN, United States) (Cite your JIR article). The expression levels were quantitated as the positively stained area per ACS using the image analysis software (WinRoof 6.5, Mitani Co. Ltd., Tokyo, Japan).

Angiogenesis was analyzed by immunostaining with rat anti-mouse CD31 monoclonal antibody (Clone 390, Biolegend Inc, San Diego, CA, United States) previously described. The neovessels number were counted under the microscope and reported as CD31-positive vessels per ACS (20, 21).

Prism 9.0 software was used for all statistical analyses. Continuous variables were expressed as mean and standard deviation (normal distribution) or media with interquartile (not normal distribution). Student's t and nonparametric Mann-Whitney tests were used for normally and nonnormally distributed data, respectively. Two-way ANOVA followed by Sidak's multiple comparisons test was used for testing statistical difference for aortic diameters among groups. Statistical significance level was set at p < 0.05.

As illustrated in Figures 1A,B, PIAS3 deficient mice were successfully generated by CRISPR/Cas9 techniques. PCR genotyping demonstrated the homozygotes of PIAS3 deficiency. qRT-PCR and Western blotting analysis further confirmed the deficiency of PIAS3 at mRNA and protein levels in PIAS3^−/− as compared to PIAS3^+/+ mice.

Intra-infrarenal aortic PPE infusion was conducted to induce AAAs in PIAS3^+/+ and PIAS3^−/− mice. Experimental AAAs were successfully induced in PIAS3^+/+ mice (Figure 1C). PIAS3 deficiency significantly inhibited aortic expansion as compared to PIAS3^+/+ mice. Aortic diameter on day 14 after PPE infusion were 1.03 ± 0.07 mm and 1.26 ± 0.10 mm for PIAS3^−/− and PIAS3^+/+ mice, respectively (Figure 1C). After subtracting an average aortic dilation caused by PBS infusion (approximately 0.8 mm), PIAS3 deficiency reduced PPE-induced aortic expansion by approximately 50% (Figure 1C, middle panel). The diameter increase over the baseline was significantly less in PIAS3^−/− than that in PIAS3^+/+ mice (Figure 1C, right panel).

Histological analysis was performed in aneurysmal aortas. In H&E staining, PPE infusion induced a remarkable aortic dilation, predominant inflammatory cell infiltration, medial elastin degradation and SMC loss (Figure 1D, left panel). However, PIAS3^−/− mice were protective against the destruction of medial elastin and SMCs. In comparison with PIAS3^+/+ mice, the integrity of medial elastin was relatively preserved in PIAS3^−/− mice, with significantly reduced elastin degradation score (Figure 1D, middle panel). Similar was true for medial SMCs (Figure 1D, right panel).

In aortic immunostaining, the score for the accumulation of macrophages, identified by CD68-positve cells, was significantly lower in PIAS3^−/− than that in PIAS3^+/+ mice (Figures 2A,B). Similarly, PIAS3 deficiency reduced the accumulation of CD4⁺ T cells, CD8⁺ T cells and B cells by approximately 50% (Figure 2).

MMP2 and MMP9 are contributors of PPE-induced AAAs (17, 22, 23). The levels of aortic MMP2 and MMP 9 were diminished in PIAS3^−/− as compared to PIAS3^+/+ mice (Figure 3A). In semi-quantitative analysis, individual MMP positively stained areas were reduced by approximately 61% and 70% for MMP2 and MMP9, respectively, in PIAS3^−/− as compared to PIAS3^+/+ mice (Figures 3B,C).

Mural angiogenesis is involved in AAA pathogenesis (24–27). We thus determined whether the protective effect of PIAS3 deficiency on AAAs was associated with altered angiogenesis. In immunostaining of CD31, a maker of angiogenesis, the neovessels were significantly less in PIAS3^−/− than that in PIAS3^+/+ mice (Figure 4, left panel), with a 40% reduction in mural neovessels in PIAS3^−/− mice (Figure 4, right panel).