CRP deficient mice were previously generated using CRISPR/Cas9 and homologous recombination technology to knock-in a STOP cassette at the ATG site of the CRP gene on C57BL/6 genetic background) at Shanghai Biomodel Organism Science & Technology Development Co., Ltd (Shanghai, China) (24). CRP deficient and C57BL/6 wild type (WT) mice were used for all experiments. The use and care of animals in this study were approved by the Laboratory Animal Management Committee of Xi’an Jiaotong University, Xi’an, China (No. 2022-623).

Genomic DNA was extracted from tail tips of less than 3 weeks old mice. Genotyping was conducted using the polymerase chain reaction (PCR) assay and gene-specific primer set. PCR primers were sense primer P1, (GCAGTTGGCCAGGGAAAGTT) and antisense primer P2 (CATGATCAGAAGGCACCAGAGTAG) for WT allele (PCR product size: 552 base pairs) and sense primer P1 and antisense primer P3 (CCTCGCCGGACACGCTGAA) for targeted allele (PCR product size: 637 base pairs), respectively. Quantitative real-time reverse transcription- polymerase chain reaction (qRT-PCR) was assessed CRP mRNA levels in liver tissues using the primers ( Table 1 ). Western blotting analysis was utilized to determine CRP protein expression. Antibodies for Western blotting were a goat anti-mouse CRP polyclonal antibody (1:4000, Cat#: BAF1829, R&D Systems, Inc, Minneapolis, MN, USA), a rabbit anti-mouse β-actin polyclonal antibody (1:10000, Cat#: bs-0061R, Bioss Technology, Beijing, China), and an horseradish peroxidase-conjugated donkey anti-goat polyclonal antibody (1:5000, Cat#: EK030, Zhuangzhi Biological Technology, Xi’an, China) or goat anti-rabbit polyclonal antibody (1:10000, Cat#: bs-40295G, Bioss Technology) (25).

Male CRP-deficient and WT control mice at 9 weeks of age were used for experiments. AAAs were induced in infrarenal aorta under sterile condition using porcine pancreatic elastase (PPE) infusion method as previously described (26–28). Briefly, mice were anesthetized by 2% isoflurane inhalation, and a laparotomy was created to expose the infrarenal aorta. Mice were infused with PPE solution for 5 minutes (1.5 units/mL, Cat#: E-1250, Sigma-Aldrich, St Louis, MO, USA) through the aortotomy in temporarily controlled infrarenal aortic segment using a pressure pump (29). Thereafter, aortotomy and laparotomy were sequentially closed using 10-0 and 6-0 silk sutures, respectively. Mice were recovered, housed in individual cages, and monitored daily for morbidity and mortality.

External infrarenal aortic diameters were measured in situ using a digital microscope. Briefly, infrarenal aortic segment was photographed immediately prior to (baseline) and 14 days following PPE. Maximal external infrarenal aortic diameters were determined using Motic Image Plus 3.0 ML software (Motic Electric Group Co., Ltd, Xiamen, China). An AAA was defined as a more than 50% increase in external diameter over the baseline level (29).

Mice were euthanized by carbon dioxide inhalation 14 days following PPE infusion. Aortas were harvested, embedded in optical cutting temperature compound, sectioned (6 μm) and fixed with cold acetone. Infrarenal aortas from naïve WT mice were processed identically and served as non-aneurysmal controls. Frozen sections were stained with hematoxylin and eosin (H&E) and a standard streptavidin peroxidase immunohistochemical method for the assessment of morphological alterations and CRP protein expression, respectively. Reagents for CRP tissue immunostaining were a rabbit anti-mouse CRP polyclonal antibody (1:200, Cat#: bs-0155R) and normal rabbit IgG (1:200, Cat#: bs-0295P) from Bioss Technology. Other reagents were biotinylated goat anti-rabbit IgG antibody (1:400, Cat#: BA-1000) and AEC substrate kit (Cat#: SK-4200, Vector Laboratories, Inc, Newark, CA, USA) and streptavidin-peroxidase conjugate (1:200, Cat#: 016-030-084, Jackson ImmunoResearch, West Grove, PA, USA).

H&E and Elastic van Gieson (EVG) staining were performed frozen aortic sections to assess general morphological changes and elastin contents, respectively. To assess SMC retention, frozen aortic sections were sequentially stained with a goat anti-SMC α-actin polyclonal antibody (1:200, Cat#: NB300-978, Novus Biologicals, Centennial, CO, USA), a biotinylated rabbit anti-goat IgG antibody (1:400, Cat#: BA-5000, Vector Laboratories, Inc) and streptavidin-peroxidase conjugate (1:200, Cat#: 016-030-084, Jackson ImmunoResearch), and staining was visualized using the AEC substrate kit (Vector Laboratories, Inc). Elastin fragmentation and SMC loss were scored as the grade I (mild) to IV (severe) using a histological grading system as reported previously (26–31).

A standard biotin-streptavidin-peroxidase method was used for all immunohistochemical staining. Reagents used in this study were monoclonal antibodies against CD68 (macrophages, 1:200, clone #: FA-11, Cat#: 137002), CD4 (CD4⁺ T cells, 1:200, clone #: GK1.5, Cat#: 100402), CD8 (CD8⁺ T cells, 1:200, clone #: 53-6.7, Cat#: 100702), B220 (B cells, 1:200, clone #: RA3-6B2, Cat#: 103202) and CD31 (1:200, clone#: 390, Cat#: 102402) (all above mentioned primary antibodies from Biolegend Inc, San Diego, CA, USA), goat anti-mouse polyclonal antibodies against MMP2 (1:200, Cat#: AF1488, R&D Systems) and MMP9 (1:200, Cat#: AF909, R&D Systems), a biotinylated goat anti-rat secondary antibody (1:400, Cat#: BA-9400, Vector Laboratories, Inc) or rabbit anti-goat IgG secondary antibody (1:400, Cat#: BA-5000, Vector Laboratories, Inc), and streptavidin-peroxidase conjugate (1:200, Cat#: 016-030-084, Jackson ImmunoResearch) (27–32). Macrophage accumulation was scored as the grade I to IV, and the densities of CD4⁺ T cells, CD8⁺ T cells, B220⁺ B cells and angiogenesis were quantified as the number of positively stained cells or neovessels per aortic cross section (ACS) (30). MMP expression levels were quantified as a positively stained area percentage of aortic wall using WinRood 6.5 image software, Mitani Co. Ltd., Tokyo, Japan.

Primary macrophages were isolated from 8 weeks old WT and CRP deficient mice 3 days following intraperitoneal injection of 3% thioglycolate and suspended in RPMI-1640 medium containing 10% fetal bovine serum and 1% penicillin/streptomycin. Following 6 hours activation with lipopolysaccharide (LPS, 50 ng/mL, Cat#: L2630, Sigma-Aldrich) or incubation with vehicle at 37°C, 5% CO₂ for 6 h, cells were harvested. For interleukin-4 (IL-4) stimulated experiment, macrophages were treated with IL-4 (10 ng/mL; Cat#: 214-14, PeproTech, Inc. Cranbury, NJ, USA) or vehicle for 16 hours. Total RNA was extracted with RNAiso Plus (Cat#: 9109), and complementary DNA was synthesized PrimeScript RT kit (Cat#: RR036A) (all from Takara Bio Inc, Kusatsu, Shiga, Japan) and amplified using RealStar Green Power mix (Cat#: A311-10; GenStar, Beijing, China) and gene-specific primers ( Table 1 ). mRNA levels were quantitated as fold changes in relative to vehicle-treated macrophages.

All statistical analyses were performed using the GraphPad software Version 9.0 (Boston, MA, USA). Data on continuous variables were presented as either mean ± standard deviation if normally distributed, and statistical significance was tested using Student’s t-test, or one, two–way ANOVA followed by two group comparison tests. Otherwise, data were given as median and interquartile range, and statistical significance was determined using nonparametric Mann-Whitney test. Statistical significance level was set at p<0.05.

To examine whether CRP protein expression is altered in aneurysmal aortas, we stained aortic frozen sections from aneurysmal (PPE-infused) and non-aneurysmal WT mice. As depicted in Figure 1 , rare or no positive staining was not seen in non-aneurysmal aorta. In contrast, an intense and diffusion CRP staining was noted in aneurysmal aortas ( Figure 1 ). In serial sections, CRP staining was coincident with inflammatory cell accumulation in aneurysmal aortas.

To clarify the effect of CRP in experimental AAAs, previously generated CRP-deficient (CRP^-/-) mice were used in this study (24). The homozygotes of CRP-deficient mice were screened by PCR genotyping and Western blotting ( Figure 2A ). Luminal PPE infusion in infrarenal aorta was performed to induce AAAs in both WT and CRP^-/- mice. Fourteen days following PPE infusion, aortic dilation, as measured by external aortic diameter, was seen in both WT and CRP^-/- mice as compared to the baseline level. However, maximal external aortic diameter was significantly smaller in CRP^-/- (1.08 ± 0.11 mm) than that in WT (1.21 ± 0.08 mm) mice on day 14 after PPE infusion ( Figures 2B, C ). When subtracting aortic dilation due to the pressed infusion (approximately 0.8 mm), CRP deficiency per se led to a 32% reduction in aneurysmal enlargement ( Figure 2C ). Even considering the influence of baseline aortic diameter, a remarkable reduction in external aortic diameter was observed in CRP^-/- as compared to WT mice 14 days following PPE infusion ( Figure 2D ). Thus, CRP may in part mediate experimental aneurysmal expansion.

In EVG staining for medial elastin, more medial elastin was maintained in CRP^-/- [score as 3 (2–3) (median with interquartile range)] as compared to WT [score as 4 (3–4)] mice ( Figures 2E, F , p=0.019). Similarly, substantial retention of SMCs was noted in CRP^-/- [score as 3 (2–3)] as compared to WT [score as 4 (3–4)]. However, the difference in SMC grades did not reach statistical significant level ( Figures 2E, G ) probably due to insufficient sample size. Thus, experimental AAA suppression mediated by CRP deficiency is associated with marked preservation of medial elastin.

Leukocytes contribute to experimental AAA pathogenesis (29, 33). In non-aneurysmal aorta, few leukocytes, almost CD68⁺ macrophages in aortic adventitia, were stained positively with subset-specific mAbs ( Figure 3A ). However, leukocytes infiltrated intensely throughout aortic wall in aneurysmal aortas ( Figure 3A ). CD68⁺ macrophages accumulated significantly less in CRP^-/- mice [score as 3 (2–4) (median with interquartile range)] than WT mice did [score as 4 (3–4)] (p=0.025) ( Figures 3A, B ). While CD4⁺ T cells, CD8⁺ T cells and B220⁺ B cells also accumulated in CRP^-/- and WT mice, no significant difference in either subset was seen between two mouse strains ( Figures 3A, C–E ).

Both MMP2 and MMP9 are important mediators of AAA pathogenesis (34). In our immunohistochemical staining, MMP2 expression was reduced in PPE-infused CRP^-/- [expressed as positive staining area ratio (%): 49.51 ± 13.77] as compared to WT (29.55 ± 9.98) mice ( Figure 4A ), with a 40% reduction in CRP^-/- mice ( Figure 4B ). However, there was no significant difference in MMP9 expression between PPE-infused CRP^-/- and WT mice ( Figures 4A, C ).

Angiogenesis, an additional key AAA pathology, also contributes to the progression of AAAs (35, 36). Angiogenesis, as determined by the density of CD31⁺ neovessels, was not differentiated between CRP^-/- (50.82 ± 13.40/ACS) and WT (43.90 ± 12.23/ACS) mice ( Figure 5 ). Thus, neoangiogenesis may have no or limited contribution to the suppression of experimental AAAs by CRP deficiency.

Classic macrophage activation (conventionally known as proinflammatory M1 macrophage activation/polarization) promotes AAA formation and progression (33, 37). In classical activated macrophage by LPS, the expression levels of mRNA for TNF-α and COX-2, but not IL-1β, IL-6 (M1 marker genes) were significantly reduced in CRP deficient as compared to WT macrophages ( Figure 6 ). However, no significant influences were found for the expression levels of all alternative activation macrophage markers (conventionally knowns as anti-inflammatory M2 macrophages) after activated by IL-4 ( Supplementary Figure 1 ).