Skip main navigation

Adipocyte-Derived Serum Amyloid A Promotes Angiotensin II–Induced Abdominal Aortic Aneurysms in Obese C57BL/6J Mice

Originally published, Thrombosis, and Vascular Biology. 2022;42:632–643



Obesity increases the risk for human abdominal aortic aneurysms (AAAs) and enhances Ang II (angiotensin II)–induced AAA formation in C57BL/6J mice. Obesity is also associated with increases in perivascular fat that expresses proinflammatory markers including SAA (serum amyloid A). We previously reported that deficiency of SAA significantly reduces Ang II–induced inflammation and AAA in hyperlipidemic apoE-deficient mice. In this study. we investigated whether adipose tissue-derived SAA plays a role in Ang II–induced AAA in obese C57BL/6J mice.


The development of AAA was compared between male C57BL/6J mice (wild type), C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 (TKO); and TKO mice harboring a doxycycline-inducible, adipocyte-specific SAA1.1 transgene (TKO-Tgfat; SAA expressed only in fat). All mice were fed an obesogenic diet and doxycycline to induce SAA transgene expression and infused with Ang II to induce AAA.


In response to Ang II infusion, SAA expression was significantly increased in perivascular fat of obese C57BL/6J mice. Maximal luminal diameters of the abdominal aorta were determined by ultrasound before and after Ang II infusion, which indicated a significant increase in aortic luminal diameters in wild type and TKO-TGfat mice but not in TKO mice. Adipocyte-specific SAA expression was associated with MMP (matrix metalloproteinase) activity and macrophage infiltration in abdominal aortas of Ang II–infused obese mice.


We demonstrate for the first time that SAA deficiency protects obese C57BL/6J mice from Ang II–induced AAA. SAA expression only in adipocytes is sufficient to cause AAA in obese mice infused with Ang II.


  • SAA (Serum amyloid A) expression in adipose tissues increases in mice infused with Ang II (angiotensin II).

  • Deficiency of all the inducible SAA isoforms protects obese mice from the development of Ang II–induced abdominal aortic aneurysms.

  • Expression of SAA only in adipose tissues drives the development of abdominal aortic aneurysm in obese C57BL/6J mice.

  • Adipose tissue-derived SAA augments the development of abdominal aortic aneurysm, possibly by recruiting macrophages and increasing MMP (matrix metalloproteinase) activity in the abdominal aorta.

Abdominal aortic aneurysms (AAAs) are a vascular disease that affects 4% to 9% of the adult male population and accounts for at least 15 000 deaths per year in the United States alone.1 The disease is typically asymptomatic, and its first presentation may be catastrophic rupture with high mortality. Unfortunately, despite decades of research, there are no treatments for this disease other than surgical procedures to replace or reinforce the dilated aortic segment.2 Therefore, it is crucial to elucidate the mechanisms involved in the development and progression of the disease.

AAA has been classified as a vascular matrix degenerative disease that involves the dilation and thinning of the artery wall in the abdominal segment of the aorta.3,4 Macrophage infiltration, inflammatory cytokine release, and MMP (matrix metalloproteinase) activation are thought to be important factors resulting in the medial injury and adventitial inflammation ultimately leading to the aneurysmal dilation of the aorta in AAA.5,6 Population-based studies indicate an association between AAA formation and increased body weight.7,8 Obesity increases the risk of cardiovascular related mortality and is a risk factor for the development of AAA in humans9 as well as in mice.10 Although the link between obesity and AAA is known, the molecular and cellular mechanisms for this link are poorly understood. There is a significant amount of perivascular adipose tissue (PVAT) accumulation surrounding the aorta, and the incidence of AAA correlates to the quantity of PVAT.11 PVAT is believed to regulate vascular biology in paracrine ways due to its anatomic proximity to the vascular wall.12 PVAT expands in obesity and can interact with inflammatory cells and vascular cells to promote vascular diseases.13

Many laboratories have shown that infusion of Ang II (angiotensin II) to hypercholesterolemic mice results in AAA formation.14,15 Male sex and obesity are risk factors for human AAA, as well as for Ang II–induced AAA in mice.10,16,17 Mice with either diet-induced or genetic obesity exhibited markedly increased inflammation in PVAT surrounding abdominal aortas and enhanced AAA formation in an Ang II–induced AAA.10 Obesity-driven AAA in this model was independent of serum cholesterol concentrations, lipoprotein distributions, changes in insulin sensitivity, and blood pressure responses to Ang II.10 In contrast, regional differences in periaortic adipocytes and their differential ability to promote chemokine release, macrophage infiltration, and proinflammatory cytokine expression was related to enhanced AAA risk in obesity.10

Acute phase SAA (serum amyloid A) is a family of secreted proteins whose concentration in the plasma increases 1000-fold or more during a systemic inflammatory response. SAA is also persistently elevated in chronic inflammatory conditions, such as diabetes,18,19 obesity,18,20 rheumatoid arthritis,21 and others. The acute phase SAAs include SAA1 and SAA2 (SAA1.1 and SAA2.1 in mice), and in mice but not humans, SAA3 is an acute phase SAA22 (in humans the SAA3 gene encodes a premature stop codon and is not expressed). A wealth of epidemiological data links SAA with cardiovascular disease and increased SAA is associated with cardiovascular disease mortality.3,23–25 While the liver is the major source of SAA during an acute phase response, SAA may also be produced in adipose tissues and its expression increases with obesity and decreases with weight loss.26 Indeed, adipocytes are thought to be a predominant source of local and even systemic SAA in the setting of obesity.20,26 The increased expression of SAA by adipocytes in obesity potentially acts as a direct link between obesity and its comorbidities, including diabetes and cardiovascular diseases.20 The goal of the current study was to determine if the expression of SAA in adipose tissue contributes to the increased AAA observed in obesity.

Materials and Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request. Details on animals and antibodies are provided in the Supplemental Major Resource Table.


All mice used in this study are in the C57BL/6J background. Mice deficient in SAA1.1, SAA2.1, and SAA3 (TKO) were generously provided by Drs June-Yong Lee and Dan Littman, New York University.27,28 SAA transgenic mice encoding an SAA transgene regulated by a tetracycline-responsive promoter were generously provided by Dr Paul Simon (University College London),29 and crossed with TKO mice. The transgenic mice in the TKO background were then crossed with mice expressing rtTA (reverse tetracycline-controlled transactivator) under control of the adipocyte-specific adiponectin promoter30 (AP-rtTA; gift from Dr Philip Scherer, University of Texas Southwestern Medical Center), which were also previously bred to the TKO background. Thus, the resulting strain (designated TKO-Tgfat) exhibit highly inducible SAA expression only in adipose tissues upon administration of doxycycline. Littermate AP-rtTA mice lacking the SAA transgene are designated TKO for simplicity throughout the text. All procedures involving animals were approved by the Institutional Animal Care and Use Committees at the University of Kentucky and the Lexington Veterans Affairs Medical Center.

Animal Treatments

Twelve- to 15-week-old male C57BL/6J mice (wild type [WT]), C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 (TKO), and transgenic mice expressing SAA only in adipose tissues when given doxycycline (TKO-TGfat mice) were fed an obesogenic diet (60% kcal from fat; D12492, Research diets) for 12 or 14 weeks, as indicated. For the 14-week AAA study, mice (n=12–15/genotype) were provided water containing 0.4 mg/mL doxycycline (Sigma D 9891) and 5% sucrose ad libitum for the past 6 weeks to induce SAA transgene expression. All experiments were performed on male mice as they develop aneurysms at a significantly higher incidence than female mice, similar to the clinical observation that AAAs predominantly affect men.31 The mice were infused with Ang II (1000 ng/kg per minute; 4006473, BACHEM) via Alzet osmotic minipumps (model 2004; Durect Corporation) for the past 4 weeks to induce AAA. Body weight was measured weekly and body composition was measured by nuclear magnetic resonance spectroscopy (EchoMRI). All mice that died during the course of Ang II infusion underwent necropsy to confirm that cause of death was due to aortic rupture.

Quantification of AAA

Ultrasound measurements were performed in mice before and after 28 days of Ang II infusion. Abdominal aortas were visualized in mice anesthetized with 2% v/v Isofluorane with medical grade oxygen (Butler Schein) using high frequency ultrasound (Vevo 660; VisualSonics) as described previously.32 Briefly, anesthetized mice were restrained in a supine position for ultrasonography, and short axis scans of abdominal aortas from the level of the left renal arterial branch moving vertically to the suprarenal region were obtained. Cine loops of 100 frames were acquired throughout the renal region of abdominal aortas. Maximal luminal diameter and area measurements were determined by 2 different observers blinded to groups. AAA incidence in treatment groups was defined as the sum of the number of mice that died from aortic rupture plus the number of mice that showed 50% or greater increase in maximal luminal diameter of the abdominal aorta after 28-day Ang II infusion compared with baseline.

Human Magnetic Resonance Imaging

Magnetic resonance imaging of the visceral cavity of lean and obese individuals (anonymized images; IRB exempt) were visualized by T1 weighed spoiled 3-dimensional gradient echo Dixon fat image with VIBE DIXON in an axial cut through the abdominal aorta.

Plasma SAA Measurements

Plasma SAA concentrations were determined using a mouse SAA ELISA kit (catalog no: TP 802M, Tridelta Development Ltd).

Plasma Cholesterol Measurements

Plasma cholesterol concentrations were measured using enzymatic kits (Wako Chemicals).


Immunohistochemistry was performed as described previously.33 Briefly, sequential paraffin-embedded sections of aortae (5-μm thick) were collected and mounted on microscope slides. Paraffin sections were deparaffinized and rehydrated through standard procedures. After antigen retrieval using Target Retrieval solution citrate buffer, pH6.0 (S2369, Dako) and quenching of endogenous peroxidase activity using 3% H2O2, sections were blocked in normal blocking serum which was prepared from the species in which the secondary antibody was made. After blocking endogenous biotin using Avidin/Biotin blocking kit (SP-Z001, Vector Laboratories), sections were incubated with rabbit anti-mouse SAA33 (1:500 dilution) overnight at 4 °C, followed by biotinylated secondary antibody for 1 hour at room temperature. Using ABC system (PK-6101, Vector Laboratories), and DAB detection system (K3468, Dako North America Inc), signals were detected. For immunofluorescence staining, aortae were frozen in optimal cutting temperature compound (4583, optimal cutting temperature; Tissue-Tek), and 8-μm thick sections down the length of the aorta were mounted on glass slides. All tissue sections were subjected to identical processing at the same time to allow for direct comparison. Sections were fixed in 4% paraformaldehyde for 30 minutes and treated with 0.1% Triton X-100 in PBS for 15 minutes. After blocking in 1% BSA/PBS at room temperature for 2 hours, slides were incubated overnight at 4 °C with a combination of rabbit anti-mouse SAA (1:200; ab199030, Abcam) and rat anti-mouse CD68 (1:200; ab53444, Abcam). After washes with PBS, SAA was detected using Alexa Fluor 568–labeled goat anti-rabbit IgG (1:200; A11011, Thermo Fischer Scientific) and CD68 was detected using Alexa Fluor 488–labeled goat anti-rat IgG (1:200; A11006, Thermo Fischer Scientific). Slides were mounted using fluorescence-protecting medium containing DAPI (Vectashield; Vector Laboratories). Images were captured by fluorescence microscopy (Nikon Eclipse 80i microscope, Nikon Instruments) and quantified using Nikon NIS-elements software.

For elastin staining, optimal cutting temperature-embedded sections were fixed in 10% formalin and treated according to manufacturer’s instructions (Elastic Stain Kit, HT25A1KT, Thermo Fisher Scientific). Images were captured on a Nikon ECLIPSE 80i microscope with the aid of NIS-Elements BR 4.00.08 software.

In Situ Zymography

In situ zymography was performed as described earlier.33 Briefly, optimal cutting temperature-embedded abdominal aorta sections, adjacent to those used for immunohistochemistry (see above), were incubated with 20 μg/mL DQ gelatin fluorescein conjugate for 2 hours at 37 °C according to kit instructions (EnzChek Gelatinase/Collagenase Assay Kit, E-12055, Molecular Probes, Inc). The fluorescence generated by hydrolysis of the added substrate was recorded by an Olympus IX70 microscope equipped with Olympus DP70 digital camera. The general MMP inhibitor 1,10-phenanthroline, 20 mmol/L, was used to define nonspecific fluorescence (33510, Sigma-Aldrich).

Adipocyte Area Measurement

Epididymal fat was collected and fixed in 10% formalin, paraffin embedded, cut into 5-μm sections, and stained with hematoxylin (Vector Laboratories, Burlingame, CA). Average adipocyte area was determined from 4 randomly chosen frames from 4 adipose tissue sections of 4 mice of each group using Nikon NIS-elements software.

RNA Isolation and Quantitative Real-Time Polymerase Chain Reaction

Total RNA was isolated from mouse adipose tissues according to the manufacturer’s instructions (RNeasy Mini Kit, 74106, Qiagen). RNA samples were incubated with DNase I (79254, Qiagen) for 15 min at RT before reverse transcription. Adipose tissue RNA (0.5 μg) was reverse transcribed into cDNA using the Reverse Transcription System (4368814, Applied Biosystems). After 4-fold dilution, 5 µl was used as a template for real-time polymerase chain reaction. Amplification was done for 40 cycles using Power SYBR Green Polymerase Chain Reaction master Mix Kit (4367659, Applied Biosystems). Quantification of mRNA was performed using the ΔΔCT method and normalized to GAPDH. Primer sequences will be provided on request.

Western Blotting

Liver and adipose tissue homogenates were prepared by homogenizing tissues (50–100 mg) in RIPA buffer (R0278, Sigma) containing protease inhibitor cocktail (88665, Thermo Fisher). Aliquots corresponding to 25 µg of protein were separated on a 4% to 20% polyacrylamide gradient gel (5671095, Bio-Rad) and immunoblotted with anti-SAA primary antibody (1 ng/mL, ab199030, Abcam). The secondary antibody (anti-rabbit antibody; 1:10 000 dilution; ab205718) was from Abcam. For loading controls, liver and adipose tissue lysates/membranes were immunoblotted with anti-β actin primary antibody (1:2000; Sigma, A5441) and anti-mouse secondary antibody (1:10 000 dilution; Sigma, A-4416).

Statistical Analyses

Statistical analyses were performed using either SigmaPlot software (Version 14.0) or GraphPad Prism 8. Normality and homogeneous variation were tested in all continuous variables by Shapiro-Wilk and Brown-Forsythe tests, respectively. Since the nonrepeated continuous data did not pass either test, all data were analyzed using nonparametric methods. For 2-group comparisons, Mann-Whitney U test was performed. To compare multiple groups, Kruskal-Wallis 1-way ANOVA on ranks followed by Dunn’s method was used. For multiple groups with repeated measures, 2-way repeated measures ANOVA followed by Bonferroni post hoc analysis. For multiple groups with 2 factors, 2-way ANOVA with Sidak multiple comparisons test was used. P<0.05 was considered statistically significant. Fisher exact test was applied to the comparisons of AAA incidence.


PVAT Is Increased in Obesity and Expresses SAA

Magnetic resonance imaging of the abdomen of humans shows negligible adipose tissue mass surrounding the aorta of a lean individual, however, there is considerable accumulation of adipose tissues surrounding the aorta of an obese individual (Figure 1A). Notably, PVAT mass of C57BL/6J mice fed an obesogenic diet for 12 weeks was significantly heavier than that of mice fed a standard diet for the same period of time (Figure S1), consistent with previous reports addressing the impact of an obesogenic diet on PVAT mass in mice.13 PVAT from obese C57BL/6J mice infused either with saline or Ang II for 28 days was analyzed for SAA expression by immunohistochemistry and quantitative real-time polymerase chain reaction. SAA immunostaining was more pronounced in the PVAT surrounding the abdominal aortas of Ang II–infused mice compared with saline-infused controls (Figure 1B), and there was a significant 4.5-fold increase in SAA1.1/2.1 mRNA abundance in the PVAT of obese mice infused with Ang II compared with saline (Figure 1C). There was a ≈3-fold increase in SAA3 mRNA abundance in the PVAT of obese mice infused with Ang II compared with saline (Figure 1C). Thus, one consequence of Ang II infusion in obese mice is an increase in SAA expression in PVAT surrounding the abdominal aorta, a region susceptible to dilation and AAA formation.

Figure 1.

Figure 1. Perivascular fat is increased in obesity and expresses SAA (serum amyloid A). A, Cross-section magnetic resonance imaging (MRI) images of the visceral cavity reveals profound increase in adipose tissue mass surrounding the aorta (*aorta) in obese compared with lean individual. Periaortic fat is indicated by the arrows. The left is from a 51-y-old male weighing 65.6 kg and the right from a 51-y-old male weighing 135 kg. B, SAA immunostaining in perivascular tissue in obese C57BL/6J male mice infused with saline or Ang II (angiotensin II), shown at 4× (upper), scale bar=200 µm and 20× (lower), scale bar=50 µm. The images are representative of 3 sections from 2 mice treated with either saline or Ang II (C) SAA 1.1/2.1 and SAA3 mRNA abundance in perivascular adipose tissue from saline and Ang II–infused obese mice (n=3–6 mice/group). Data are mean±SEM; statistical tests used: Mann-Whitney test.

TKO-Tgfat Mice Exhibit Adipose Tissue-Specific SAA Expression

To investigate whether adipose-derived SAA contributes to the formation of AAA, transgenic mice expressing SAA only in adipose tissue when given doxycycline in the drinking water were generated as described in Materials and Methods (TKO-Tgfat mice). Littermate mice lacking the SAA transgene (TKO) and also administered doxycycline were used as control in our studies. Male C57BL/6J (WT), TKO, and TKO-Tgfat mice were fed an obesogenic diet for 14 weeks to induce obesity. All mice were given doxycycline (0.4 mg/mL) in drinking water during the past 6 weeks of obesogenic diet and infused with Ang II (1000 ng/kg per minute) during the past 4 weeks (starting 2 weeks after the doxycycline treatment; Figure 2A). At the end of the study, SAA1.1 expression was significantly higher in PVAT of TKO-Tgfat mice compared with WT (Figure 2B). Consistent with lack of endogenous SAAs in TKO mice, SAA3 mRNA was only detected in PVAT from WT mice (Figure 2B). SAA1.1 expression was also significantly increased in other adipose tissue depots of TKO-Tgfat mice compared with WT mice (expression in gonadal adipose tissues shown in Figure S2). Immunoblot analysis of adipose and liver extracts confirmed SAA protein expression in adipose tissues, but not liver, of TKO-Tgfat mice (Figure 2C). In WT mice, relatively modest SAA expression was detected in both adipose tissue and liver as expected (Figure 2C). Plasma SAA levels were significantly higher in the WT mice compared with TKO-Tgfat mice (113±22 versus 27.6±6 µg/mL; Figure 2D), despite higher adipose tissue SAA expression in TKO-Tgfat mice (compare Figure 2B and 2C with 2D). As expected, SAA was not detected in the adipose tissues, liver or plasma of TKO mice (Figure 2C and 2D). Taken together, our results suggest that relative to WT mice which express moderate amounts of SAA in both hepatic and adipose tissues, TKO-Tgfat mice express SAA only in adipose tissue, which despite very high levels does not significantly contribute to the systemic pool.

Figure 2.

Figure 2. TKO-TgFat mice exhibit adipose tissue-specific SAA (serum amyloid A) expression. A, Study design: 3 groups of 12–15 wk old male mice (wild type [WT], C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 [TKO], and TKO-TgFat) were fed obesogenic diet (high-fat diet [HFD]) for a total of 14 wk. For the past 6 wk, all mice were given doxycycline-supplemented water (0.4 mg/mL in 5% sucrose water) to induce transgenic fat-specific overexpression of SAA1.1 in TKO-TgFat mice. All mice were infused with Ang II (angiotensin II; 1000 ng/kg bodyweight/min) for the past 4 wk and mice were killed at the end of 14 wk. B, SAA1.1 and SAA3 mRNA abundance in periaortic adipose tissues from Ang II–infused obese WT, TKO, and TKO-Tgfat mice. C, Adipose tissue and liver lysates (50 µg protein/lane) obtained from obese WT, TKO, and TKO-Tgfat mice at the end of the study were immunoblotted for SAA (top) and β-actin as loading control (bottom). +C is a positive control of 44 ng lipid-free SAA. D, SAA levels in plasma collected from Ang II–infused obese WT, TKO, and TKO-Tgfat mice were determined by ELISA. Data are mean±SEM. Statistical tests used: Mann-Whitney test for (B); Kruskal-Wallis test (D).

Adipocyte-Specific SAA Overexpression in Obese Mice Does Not Affect Body Weight, Body Composition, Plasma Cholesterol Levels, or Inflammatory Markers Compared With SAA-Deficient Mice

There was no significant difference in body weights between C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 (TKO) and TKO-Tgfat mice during the 14-week study. However, the WT mice gained less weight compared with TKO and TKO-Tgfat mice in response to obesogenic diet feeding (Figure 3A). Thus, in the TKO background, in the absence of endogenous SAA, the presence or absence of SAA in adipose tissues did not impact diet-induced weight gain. All groups lost weight following pump implantation as previously observed in the Ang II infusion model,27 but weight loss did not differ among groups. Transgenic overexpression of SAA in adipose tissues did not significantly change percentage body fat mass (40.3±1.0%) or lean mass (56.9±0.9%) compared with TKO mice (39.2±2.5% fat and 58.0±2.4% lean) as determined by EchoMRI after 10 weeks of diet, although there was a modest but significant increase in percentage body fat content in both TKO mice and TKO-Tgfat mice compared with the WT mice (32.3±1.3; Figure 3B). Average adipocyte size measured in gonadal adipose tissue was not significantly different among the 3 groups of mice, although there was a trend for decreased average adipocyte size in WT mice compared with TKO mice and TKO-Tgfat mice (Figure 3C). There were no apparent differences in the expression of UCP-1 (uncoupling protein-1) or perilipin1 protein, markers of brown/beige and white adipocytes respectively, in the adipose tissues of the 3 groups of mice (Figure S3A through S3C). Consistently, UCP-1 mRNA levels did not significantly differ among the adipose tissues of the 3 groups of mice (Figure S3D). Adipose tissue adiponectin and leptin mRNA expression levels were not significantly different among groups (Figure S3E and S3F). However, plasma adiponectin levels were significantly lower in TKO-Tgfat mice compared with WT and TKO mice (Figure S3G). Plasma leptin levels were not significantly different among the 3 groups of mice (Figure S3H). The expression of MMPs, MMP2, MMP9, and MMP12 (Figure S4A through S4C) as well as proinflammatory markers TNF (tumor necrosis factor)-α, IL (interleukin)-1β, F4/80, MCP (monocyte chemoattractant protein)-1 and IL-6 and anti-inflammatory markers arginase and IL-10 mRNAs in PVAT were not significantly different among the 3 groups of mice (Figure S5A through S5G). Plasma IL-6 levels were also not significantly different for the 3 strains (Figure S5H). IL-1β and TNF-α (tumor necrosis factor-α) were undetectable in plasma of all groups. Plasma total cholesterol levels were not significantly different among the 3 strains of mice at the end of the study (Figure 3D).

Figure 3.

Figure 3. Adipocyte-specific SAA (serum amyloid A) overexpression in obese mice does not affect body weight, body composition, adipocyte morphology, or plasma cholesterol levels when compared with SAA-deficient mice. A, Body weights during the course of obesogenic diet feeding. The arrow indicates the start of Ang II (angiotensin II) infusion. Data are mean±SEM; n=13–15; *P≤0.05 difference between wild type (WT) and TKO-Tgfat; #P≤0.05 difference between WT and C57BL/6J mice lacking SAA1.1, SAA2.1 and SAA3 (TKO); (B) percentage of lean (filled) and fat (open) mass of the experimental mice after 8 wk of obesogenic diet feeding as determined by EchoMRI. Data are mean±SEM n=13–15. C, HE stained sections from gonadal adipose tissue of WT, TKO, and TKO-Tgfat, shown at 20×, scale bar=50 µm. Average adipocyte area was determined from 4 randomly chosen frames from 4 adipose tissue sections of 4 mice of each group. D, Plasma total cholesterol levels in individual mice after 10 wk of obesogenic diet feeding and 2 wk of doxycycline administration. Two-way repeated measures ANOVA followed by Bonferroni post hoc analysis was used for statistics in A; Kruskal-Wallis 1-way ANOVA on Ranks followed by Dunn’s method and Sidak multiple comparison test were used in B and C, respectively.

Adipocyte-Derived SAA Restores Obesity-Accelerated AAA in TKO Mice

AAA incidence and severity were assessed in obese WT, C57BL/6J mice lacking SAA1.1, SAA2.1 and SAA3 (TKO), and TKO-Tgfat mice infused with Ang II for 28 days. Mice that died during the Ang II infusion were necropsied to determine cause of death. The incidence of death due to aortic rupture was similar for all groups (2/14, 3/14, and 3/13 in WT, TKO, and TKO-Tgfat mice, respectively; Figure 4A), and all deaths were due to aortic rupture. AAA development was assessed in surviving mice by measuring the maximal luminal diameter of abdominal aortas before and after 28 days of Ang II infusion using in vivo ultrasound. Before Ang II infusion, luminal diameters were not significantly different in WT (1.18±0.03 mm; n=14), TKO (1.14±0.032 mm; n=14), and TKO-Tgfat (1.14±0.04 mm; n=13) mice. Ang II infusion led to a significant increase in diameter of the abdominal aorta in WT (1.51±0.13; P<0.05) and TKO-Tgfat mice (1.52±0.13; P<0.05) but not TKO mice (1.29±0.07; Figure 4B), indicating that deficiency of SAA protected against AAA development and that the expression of SAA only in adipose tissue was sufficient to restore Ang II–induced AAA. The maximal external aortic diameter measured ex vivo had a similar trend and averaged 1.53±0.15, 1.30±0.15, and 1.52±0.21 mm for WT, TKO, and TKO-Tgfat mice, respectively. The overall AAA incidence, calculated as the sum of the number of mice that died from aortic rupture plus the number of mice showing >50% dilation in luminal diameter after 28-day Ang II infusion, was lower in TKO mice (14.3%) compared with WT mice (28.6%) and TKO-Tgfat mice (53.8%; Figure 4A) but did not reach statistical significance. The overall AAA incidence was 25% higher in TKO-Tgfat mice compared with the WT mice, however, the difference was not statistically significant (Figure 4A). A representative image from each group is shown in Figure 4C (the maximal aortic luminal diameter of the representative mouse imaged in each group is indicated by the red circle in Figure 4B).

Figure 4.

Figure 4. Expression of SAA (serum amyloid A) only in adipose tissues restores Ang II (angiotensin II)–induced abdominal aortic aneurysm (AAA) in obese C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 (TKO) mice. A, Mice that died during the study due to aortic rupture are represented by the solid black sections of the bars; surviving mice that developed AAA, defined as a >50% increase in luminal diameter of the abdominal aorta, are indicated by the slashed region of the bars. B, Abdominal aortas were assessed by in vivo ultrasound before (open symbols) and after Ang II infusion (filled symbols) to determine maximal luminal diameters in individual mice. Data are mean±SEM. C, Representative ex vivo images of aortas from wild type (WT), TKO and TKO-Tgfat mice are shown. The maximal luminal diameter for the corresponding ex vivo images are shown by pink circles in B. The maximal luminal diameter for the corresponding immunofluorescence images in Figure 5 are shown by green circles in B. Fischer exact test was applied to the comparisons of AAA incidence in A; Statistical significance was evaluated using 2-way ANOVA with Sidak multiple comparison test in B.

Adipocyte-Specific SAA Colocalizes With MMP Activity and Macrophage Infiltration in Abdominal Aortas of Ang II–Infused Obese Mice

AAA tissue sections (the maximal aortic diameter of the mouse imaged in each group is indicated by the green circle in Figure 4B) were examined for elastin breaks, macrophage infiltration and MMP activity, well-documented features of AAA.34 Consistent with the presence or absence of AAA, Ang II infusion resulted in elastin breaks in WT (Figure 5A, asterisk) and TKO-Tgfat (Figure 5C, asterisk) mice but not in TKO mice (Figure 5B). In WT mice, regions containing elastin breaks were distinguished by intense MMP activity visualized by in situ zymography (green fluorescence) and prominent macrophage (green fluorescence) and SAA (red fluorescence) immunoreactivity. In addition to the medial region, SAA immunostaining was also detected to a lesser extent in periaortic fat of WT mice. In TKO-Tgfat mice, regions of elastin breaks similarly contained pronounced MMP activity and macrophage immunoreactivity (Figure 5C), but SAA immunoreactivity was detected mainly in the surrounding adipose tissue and not in the aortic wall (Figure 5C). MMP activity and macrophage immunostaining was less pronounced in TKO mice as expected, as there was not an AAA. Immunostaining with the anti-SAA antibody produced only faint and diffuse background reactivity in TKO mice, demonstrating the specificity of SAA immunostaining (Figure 5B).

Figure 5.

Figure 5. Expression of SAA (serum amyloid A) only in adipocytes increases MMP (matrix metalloproteinase) activity and macrophages in the medial region of abdominal aortas in Ang II (angiotensin II)–infused obese mice. Sections showing the abdominal aortic aneurysm (AAA) and surrounding perivascular fat from wild type (WT; A), C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 [TKO] (B), and TKO-Tgfat (C) mice were processed as described in Methods to detect elastin fibers (black staining, elastin breaks are indicated by *), MMP activity (green fluorescence), macrophages (green fluorescence), and SAA (red fluorescence), as indicated. For in situ zymography and immunostaining, nuclei were identified using DAPI (blue fluorescence). Images photographed under 4× and 20× objective magnification are shown; scale bar in 4× image is 200 µm, scale bar in 20× image is 50 µm. For WT and TKO-Tgfat mice, 18 sections from individual mouse (maximal luminal diameter of the corresponding mouse in each group is indicated by green circles in Figure 4B) were processed to identify regions of elastin breaks; the corresponding region of the abdominal aorta from a representative TKO mouse is shown for comparison.


Identifying the mechanisms underlying AAA formation and progression is critical for the development of therapeutics that inhibit their expansion. Currently, there is no effective drug therapy available for preventing aneurysm progression or rupture. Recent studies suggested that targeting dysfunctional adipose tissue surrounding the vascular wall might be a useful strategy to prevent AAA rupture.17 Excessive accumulation of inflamed, dysfunctional PVAT has been proposed to be a major risk factor for endothelial dysfunction and vascular diseases, such as atherosclerosis and AAA.35

The current study expands results from our previous report that SAA is essential for the development of Ang II–induced AAA formation in hyperlipidemic apoE−/− mice.33 The major findings of the present study are that obesity-accelerated AAA in normolipidemic mice is significantly reduced in the absence of total body SAA expression, and SAA expression only in adipose tissues is sufficient to trigger the development of AAA in the setting of obesity. Obesity generates a state of low-grade inflammation characterized by an increase in the infiltration of macrophages and inflammatory factors in adipose tissues.36–38 PVAT exhibits inflammation similar to other white adipose depots with obesity.10,39,40 Our study is consistent with other studies39,40 in demonstrating that obesity increases the size of PVAT. Although SAA is thought to be primarily produced by the liver during an acute phase response, adipocytes become the predominant source of SAA with obesity,20,26 and here we show that Ang II infusion further increases SAA expression in periaortic adipose tissue of obese mice. In obese TKO-Tgfat mice that express SAA only in adipose tissues, plasma SAA levels were modest and did not reach the levels observed in obese WT mice after Ang II infusion (Figure 2D). However, the expression of SAA in adipose tissues was dramatically higher in the TKO-Tgfat mice compared with the WT mice (Figure 2C). Thus, it appears that SAA expression in adipose tissue is sufficient to exacerbate the development of obesity-driven AAA. However, while there was an ≈80× increase in SAA1.1 levels in the PVAT of TKO-Tgfat mice compared with the WT mice, this increase did not produce an additional increase in aortic diameter (Figure 4B). The lack of further AAA expansion in TKO-Tgfat mice compared with WT may reflect either a threshold effect, or contributions of systemic SAA in WT mice on AAA progression in addition to adipose tissue-derived SAA. Although a role for other adipose tissue depots cannot be ruled out, PVAT surrounding the abdominal aorta seems to be the most likely source of SAA that enhanced AAA in our studies, given the low levels of circulating SAA in obese TKO-Tgfat mice.

Whether liver-derived systemic SAA can act on the vasculature to exacerbate AAA requires further study. We have determined, for example, that one consequence of Ang II infusion in hyperlipidemic mice is upregulation of SAA in the liver and increased plasma SAA. Ongoing studies in our laboratory are investigating whether liver-derived SAA is sufficient for Ang II–induced AAA in hyperlipidemic mice.

Our data provide some insights into mechanisms by which SAA contributes to Ang II–induced AAA in obese mice. There were no significant differences in plasma cholesterol in WT, C57BL/6J mice lacking SAA1.1, SAA2.1, and SAA3 (TKO), and TKO-Tgfat mice, indicating that adipose-derived SAA’s effect on AAA was not driven by inducing hyperlipidemia. There were modest but significant differences in body weight gain and percentage fat between WT and TKO and between WT and TKO-Tgfat mice (Figure 3A and 3B); however, TKO and TKO-Tgfat mice did not show any difference in body weight gain or fat content, despite the differences in AAA formation between these groups. SAA is known to promote inflammation by acting as a chemokine and by upregulating inflammatory mediators, including MMPs,41–44 that have been implicated in AAA.34 MMP activities, macrophages, and elastin breaks appeared to be absent in abdominal aortas of Ang II–infused TKO mice lacking SAA, whereas those were readily detectable in WT and transgenic mice with adipocyte-specific SAA expression (Figure 5). These data are in line with our previous study in hyperlipidemic mice.33 It is notable that the low rupture rate (≈14%–23%) was similar for the 3 strains of mice with Ang II infusion, despite significantly increased dilation of the abdominal aorta of surviving WT and TKO-Tgfat mice compared with TKO mice at the end of the 28-day infusion (Figure 4A). Ex vivo observation of thoracic regions of the aortas in these mice did not indicate any overt differences in appearance among the 3 strains of mice. Further studies are needed to investigate the interesting possibility that SAA plays little/no role in the initiating events leading to AAA rupture, but contributes to pathological remodeling during AAA progression.

In TKO-Tgfat mice, SAA staining appeared to be visible primarily in PVAT surrounding AAA tissue, with little/no evidence of SAA in the media. Thus, the enhanced AAA formation in TKO-Tgfat mice does not appear to be the result of the direct action of SAA on vascular cells, but possibly by indirect signaling events triggered by adipocyte-derived SAA. In contrast, sections from WT mice revealed colocalization of SAA with macrophages in AAA tissue (Figure 5A). Whether this SAA is deposited from the circulation or expressed by infiltrating macrophages is not currently known. The exact mechanism of how SAA in adipocytes signals to enhance macrophage infiltration and MMP activities in the adjacent aorta remains to be established. The increased MMP activity in obese TKO-Tgfat mice could be mediated at least, in part, by the increased macrophage content. In our earlier study27 in hyperlipidemic apoE-deficient mice, we showed that whole-body deficiency of SAA was associated with significantly blunted MMP2 expression and activity in abdominal aortas of Ang II–infused mice, and concluded that SAA mediates Ang II–induced AAA in part by promoting elastin degradation.33 There was a significant reduction in plasma adiponectin levels in TKO-Tgfat compared with both WT and TKO mice consistent with the previous reports that there is an inverse relationship between SAA and adiponectin expression.45,46 Adiponectin suppresses AAA formation in apoE-deficient hyperlipidemic mice.47 Whether adiponectin plays any role in SAA-mediated AAA formation in the current study needs to be investigated. There were no significant changes in adipocyte size and leptin levels among the groups of mice. It is unclear at this point whether the mechanisms by which SAA regulates hyperlipidemia-induced AAA are different or the same as that of obesity-induced AAA.

In summary, we demonstrated that TKO mice are strikingly protected against the development of Ang II–induced AAA in obese mice, indicating that SAA is required for AAA formation in the setting of obesity. SAA expression only in adipose tissue, despite very low systemic circulating levels, is sufficient to induce AAA to a similar extent as seen in obese WT mice. This implies that SAA is exerting paracrine effects and that the accumulation of perivascular fat in obesity may be a direct cause of AAA. Studies to evaluate whether the suppression of SAA either systemically, or in PVATs, can attenuate AAA progression are urgently needed.

Article Information


We are grateful to Dr Maria C de Beer for helpful discussions and support throughout the study.

Supplemental Material

Figures S1–S5

Nonstandard Abbreviations and Acronyms


abdominal aortic aneurysm

Ang II

angiotensin II




monocyte chemoattractant protein 1


matrix metalloproteinase


perivascular adipose tissue


reverse tetracycline-controlled transactivator


serum amyloid A


tumor necrosis factor


wild type

Disclosures None.


This manuscript was sent to William C. Sessa, Senior Consulting Editor, for review by expert referees, editorial decision, and final disposition.

Supplemental Material is available at

For Sources of Funding and Disclosures, see page 642.

Correspondence to: Preetha Shridas, PhD, Department of Internal Medicine, University of Kentucky, 537 CT Wethington Bldg, 900 S Limestone St, Lexington, KY. Email


  • 1. Upchurch GR, Schaub TA. Abdominal aortic aneurysm.Am Fam Physician. 2006; 73:1198–1204.MedlineGoogle Scholar
  • 2. Lederle FA. Abdominal aortic aneurysm: still no pill.Ann Intern Med. 2013; 159:852–853. doi: 10.7326/0003-4819-159-12-201312170-00012CrossrefMedlineGoogle Scholar
  • 3. Johnson BD, Kip KE, Marroquin OC, Ridker PM, Kelsey SF, Shaw LJ, Pepine CJ, Sharaf B, Bairey Merz CN, Sopko G, et al; National Heart, Lung, and Blood Institute. Serum amyloid A as a predictor of coronary artery disease and cardiovascular outcome in women: the National Heart, Lung, and Blood Institute-Sponsored Women’s Ischemia Syndrome Evaluation (WISE).Circulation. 2004; 109:726–732. doi: 10.1161/01.CIR.0000115516.54550.B1LinkGoogle Scholar
  • 4. Nordon IM, Hinchliffe RJ, Loftus IM, Thompson MM. Pathophysiology and epidemiology of abdominal aortic aneurysms.Nat Rev Cardiol. 2011; 8:92–102. doi: 10.1038/nrcardio.2010.180CrossrefMedlineGoogle Scholar
  • 5. Daugherty A, Powell JT. Recent highlights of ATVB: aneurysms.Arterioscler Thromb Vasc Biol. 2014; 34:691–694. doi: 10.1161/ATVBAHA.114.303353LinkGoogle Scholar
  • 6. Davis FM, Rateri DL, Daugherty A. Mechanisms of aortic aneurysm formation: translating preclinical studies into clinical therapies.Heart. 2014; 100:1498–1505. doi: 10.1136/heartjnl-2014-305648CrossrefMedlineGoogle Scholar
  • 7. Agmon Y, Khandheria BK, Meissner I, Schwartz GL, Sicks JD, Fought AJ, O’Fallon WM, Wiebers DO, Tajik AJ. Is aortic dilatation an atherosclerosis-related process? Clinical, laboratory, and transesophageal echocardiographic correlates of thoracic aortic dimensions in the population with implications for thoracic aortic aneurysm formation.J Am Coll Cardiol. 2003; 42:1076–1083. doi: 10.1016/s0735-1097(03)00922-7CrossrefMedlineGoogle Scholar
  • 8. Alcorn HG, Wolfson SK, Sutton-Tyrrell K, Kuller LH, O’Leary D. Risk factors for abdominal aortic aneurysms in older adults enrolled in The Cardiovascular Health Study.Arterioscler Thromb Vasc Biol. 1996; 16:963–970. doi: 10.1161/01.atv.16.8.963LinkGoogle Scholar
  • 9. Wang L, Djousse L, Song Y, Akinkuolie AO, Matsumoto C, Manson JE, Gaziano JM, Sesso HD. Associations of diabetes and obesity with risk of abdominal aortic aneurysm in men.J Obes. 2017; 2017:3521649. doi: 10.1155/2017/3521649CrossrefMedlineGoogle Scholar
  • 10. Police SB, Thatcher SE, Charnigo R, Daugherty A, Cassis LA. Obesity promotes inflammation in periaortic adipose tissue and angiotensin II-induced abdominal aortic aneurysm formation.Arterioscler Thromb Vasc Biol. 2009; 29:1458–1464. doi: 10.1161/ATVBAHA.109.192658LinkGoogle Scholar
  • 11. Lehman SJ, Massaro JM, Schlett CL, O’Donnell CJ, Hoffmann U, Fox CS. Peri-aortic fat, cardiovascular disease risk factors, and aortic calcification: the Framingham Heart Study.Atherosclerosis. 2010; 210:656–661. doi: 10.1016/j.atherosclerosis.2010.01.007CrossrefMedlineGoogle Scholar
  • 12. Aghamohammadzadeh R, Withers S, Lynch F, Greenstein A, Malik R, Heagerty A. Perivascular adipose tissue from human systemic and coronary vessels: the emergence of a new pharmacotherapeutic target.Br J Pharmacol. 2012; 165:670–682. doi: 10.1111/j.1476-5381.2011.01479.xCrossrefMedlineGoogle Scholar
  • 13. Kim HW, Shi H, Winkler MA, Lee R, Weintraub NL. Perivascular adipose tissue and vascular perturbation/atherosclerosis.Arterioscler Thromb Vasc Biol. 2020; 40:2569–2576. doi: 10.1161/ATVBAHA.120.312470LinkGoogle Scholar
  • 14. Daugherty A, Cassis L. Chronic angiotensin II infusion promotes atherogenesis in low density lipoprotein receptor -/- mice.Ann N Y Acad Sci. 1999; 892:108–118. doi: 10.1111/j.1749-6632.1999.tb07789.xCrossrefMedlineGoogle Scholar
  • 15. Daugherty A, Manning MW, Cassis LA. Angiotensin II promotes atherosclerotic lesions and aneurysms in apolipoprotein E-deficient mice.J Clin Invest. 2000; 105:1605–1612. doi: 10.1172/JCI7818CrossrefMedlineGoogle Scholar
  • 16. Henriques T, Zhang X, Yiannikouris FB, Daugherty A, Cassis LA. Androgen increases AT1a receptor expression in abdominal aortas to promote angiotensin II-induced AAAs in apolipoprotein E-deficient mice.Arterioscler Thromb Vasc Biol. 2008; 28:1251–1256. doi: 10.1161/ATVBAHA.107.160382LinkGoogle Scholar
  • 17. Kugo H, Tanaka H, Moriyama T, Zaima N. Pathological implication of adipocytes in AAA development and the rupture.Ann Vasc Dis. 2018; 11:159–168. doi: 10.3400/avd.ra.17-00130CrossrefMedlineGoogle Scholar
  • 18. Leinonen E, Hurt-Camejo E, Wiklund O, Hultén LM, Hiukka A, Taskinen MR. Insulin resistance and adiposity correlate with acute-phase reaction and soluble cell adhesion molecules in type 2 diabetes.Atherosclerosis. 2003; 166:387–394. doi: 10.1016/s0021-9150(02)00371-4CrossrefMedlineGoogle Scholar
  • 19. Ebeling P, Teppo AM, Koistinen HA, Viikari J, Rönnemaa T, Nissén M, Bergkulla S, Salmela P, Saltevo J, Koivisto VA. Troglitazone reduces hyperglycaemia and selectively acute-phase serum proteins in patients with Type II diabetes.Diabetologia. 1999; 42:1433–1438. doi: 10.1007/s001250051315CrossrefMedlineGoogle Scholar
  • 20. Yang RZ, Lee MJ, Hu H, Pollin TI, Ryan AS, Nicklas BJ, Snitker S, Horenstein RB, Hull K, Goldberg NH, et al. Acute-phase serum amyloid A: an inflammatory adipokine and potential link between obesity and its metabolic complications.PLoS Med2006; 3:e287. doi: 10.1371/journal.pmed.0030287CrossrefMedlineGoogle Scholar
  • 21. Wong M, Toh L, Wilson A, Rowley K, Karschimkus C, Prior D, Romas E, Clemens L, Dragicevic G, Harianto H, et al. Reduced arterial elasticity in rheumatoid arthritis and the relationship to vascular disease risk factors and inflammation.Arthritis Rheum. 2003; 48:81–89. doi: 10.1002/art.10748CrossrefMedlineGoogle Scholar
  • 22. Tannock LR, De Beer MC, Ji A, Shridas P, Noffsinger VP, den Hartigh L, Chait A, De Beer FC, Webb NR. Serum amyloid A3 is a high density lipoprotein-associated acute-phase protein.J Lipid Res. 2018; 59:339–347. doi: 10.1194/jlr.M080887CrossrefMedlineGoogle Scholar
  • 23. Kosuge M, Ebina T, Ishikawa T, Hibi K, Tsukahara K, Okuda J, Iwahashi N, Ozaki H, Yano H, Kusama I, et al. Serum amyloid A is a better predictor of clinical outcomes than C-reactive protein in non-ST-segment elevation acute coronary syndromes.Circ J. 2007; 71:186–190. doi: 10.1253/circj.71.186CrossrefMedlineGoogle Scholar
  • 24. Ogasawara K, Mashiba S, Wada Y, Sahara M, Uchida K, Aizawa T, Kodama T. A serum amyloid A and LDL complex as a new prognostic marker in stable coronary artery disease.Atherosclerosis. 2004; 174:349–356. doi: 10.1016/j.atherosclerosis.2004.01.030CrossrefMedlineGoogle Scholar
  • 25. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women.N Engl J Med. 2000; 342:836–843. doi: 10.1056/NEJM200003233421202CrossrefMedlineGoogle Scholar
  • 26. Poitou C, Viguerie N, Cancello R, De Matteis R, Cinti S, Stich V, Coussieu C, Gauthier E, Courtine M, Zucker JD, et al. Serum amyloid A: production by human white adipocyte and regulation by obesity and nutrition.Diabetologia. 2005; 48:519–528. doi: 10.1007/s00125-004-1654-6CrossrefMedlineGoogle Scholar
  • 27. Lee JY, Hall JA, Kroehling L, Wu L, Najar T, Nguyen HH, Lin WY, Yeung ST, Silva HM, Li D, et al. Serum amyloid A proteins induce pathogenic Th17 cells and promote inflammatory disease.Cell. 2020; 180:79–91.e16. doi: 10.1016/j.cell.2019.11.026CrossrefMedlineGoogle Scholar
  • 28. Thompson JC, Wilson PG, Shridas P, Ji A, de Beer M, de Beer FC, Webb NR, Tannock LR. Serum amyloid A3 is pro-atherogenic.Atherosclerosis. 2018; 268:32–35. doi: 10.1016/j.atherosclerosis.2017.11.011CrossrefMedlineGoogle Scholar
  • 29. Simons JP, Al-Shawi R, Ellmerich S, Speck I, Aslam S, Hutchinson WL, Mangione PP, Disterer P, Gilbertson JA, Hunt T, et al. Pathogenetic mechanisms of amyloid A amyloidosis.Proc Natl Acad Sci USA. 2013; 110:16115–16120. doi: 10.1073/pnas.1306621110CrossrefMedlineGoogle Scholar
  • 30. Wang QA, Tao C, Gupta RK, Scherer PE. Tracking adipogenesis during white adipose tissue development, expansion and regeneration.Nat Med. 2013; 19:1338–1344. doi: 10.1038/nm.3324CrossrefMedlineGoogle Scholar
  • 31. Kent KC, Zwolak RM, Egorova NN, Riles TS, Manganaro A, Moskowitz AJ, Gelijns AC, Greco G. Analysis of risk factors for abdominal aortic aneurysm in a cohort of more than 3 million individuals.J Vasc Surg. 2010; 52:539–548. doi: 10.1016/j.jvs.2010.05.090CrossrefMedlineGoogle Scholar
  • 32. Barisione C, Charnigo R, Howatt DA, Moorleghen JJ, Rateri DL, Daugherty A. Rapid dilation of the abdominal aorta during infusion of angiotensin II detected by noninvasive high-frequency ultrasonography.J Vasc Surg. 2006; 44:372–376. doi: 10.1016/j.jvs.2006.04.047CrossrefMedlineGoogle Scholar
  • 33. Webb NR, De Beer MC, Wroblewski JM, Ji A, Bailey W, Shridas P, Charnigo RJ, Noffsinger VP, Witta J, Howatt DA, et al. Deficiency of endogenous acute-phase serum amyloid A protects apoE-/- Mice from angiotensin II-Induced abdominal aortic aneurysm formation.Arterioscler Thromb Vasc Biol. 2015; 35:1156–1165. doi: 10.1161/ATVBAHA.114.304776LinkGoogle Scholar
  • 34. Lu H, Rateri DL, Bruemmer D, Cassis LA, Daugherty A. Novel mechanisms of abdominal aortic aneurysms.Curr Atheroscler Rep. 2012; 14:402–412. doi: 10.1007/s11883-012-0271-yCrossrefMedlineGoogle Scholar
  • 35. Gu P, Xu A. Interplay between adipose tissue and blood vessels in obesity and vascular dysfunction.Rev Endocr Metab Disord. 2013; 14:49–58. doi: 10.1007/s11154-012-9230-8CrossrefMedlineGoogle Scholar
  • 36. Weisberg SP, Hunter D, Huber R, Lemieux J, Slaymaker S, Vaddi K, Charo I, Leibel RL, Ferrante AW. CCR2 modulates inflammatory and metabolic effects of high-fat feeding.J Clin Invest. 2006; 116:115–124. doi: 10.1172/JCI24335CrossrefMedlineGoogle Scholar
  • 37. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW. Obesity is associated with macrophage accumulation in adipose tissue.J Clin Invest. 2003; 112:1796–1808. doi: 10.1172/JCI19246CrossrefMedlineGoogle Scholar
  • 38. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A, Ross JS, Tartaglia LA, et al. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance.J Clin Invest. 2003; 112:1821–1830. doi: 10.1172/JCI19451CrossrefMedlineGoogle Scholar
  • 39. Vela D, Buja LM, Madjid M, Burke A, Naghavi M, Willerson JT, Casscells SW, Litovsky S. The role of periadventitial fat in atherosclerosis.Arch Pathol Lab Med. 2007; 131:481–487. doi: 10.5858/2007-131-481-TROPFICrossrefMedlineGoogle Scholar
  • 40. Henrichot E, Juge-Aubry CE, Pernin A, Pache JC, Velebit V, Dayer JM, Meda P, Chizzolini C, Meier CA. Production of chemokines by perivascular adipose tissue: a role in the pathogenesis of atherosclerosis?Arterioscler Thromb Vasc Biol. 2005; 25:2594–2599. doi: 10.1161/01.ATV.0000188508.40052.35LinkGoogle Scholar
  • 41. Lee HY, Kim SD, Shim JW, Lee SY, Lee H, Cho KH, Yun J, Bae YS. Serum amyloid A induces CCL2 production via formyl peptide receptor-like 1-mediated signaling in human monocytes.J Immunol. 2008; 181:4332–4339. doi: 10.4049/jimmunol.181.6.4332CrossrefMedlineGoogle Scholar
  • 42. Song C, Hsu K, Yamen E, Yan W, Fock J, Witting PK, Geczy CL, Freedman SB. Serum amyloid A induction of cytokines in monocytes/macrophages and lymphocytes.Atherosclerosis. 2009; 207:374–383. doi: 10.1016/j.atherosclerosis.2009.05.007CrossrefMedlineGoogle Scholar
  • 43. Connolly M, Mullan RH, McCormick J, Matthews C, Sullivan O, Kennedy A, FitzGerald O, Poole AR, Bresnihan B, Veale DJ, et al. Acute-phase serum amyloid A regulates tumor necrosis factor α and matrix turnover and predicts disease progression in patients with inflammatory arthritis before and after biologic therapy.Arthritis Rheum. 2012; 64:1035–1045. doi: 10.1002/art.33455CrossrefMedlineGoogle Scholar
  • 44. Mullan RH, Bresnihan B, Golden-Mason L, Markham T, O’Hara R, FitzGerald O, Veale DJ, Fearon U. Acute-phase serum amyloid A stimulation of angiogenesis, leukocyte recruitment, and matrix degradation in rheumatoid arthritis through an NF-kappaB-dependent signal transduction pathway.Arthritis Rheum. 2006; 54:105–114. doi: 10.1002/art.21518CrossrefMedlineGoogle Scholar
  • 45. Frühbeck G, Catalán V, Rodríguez A, Ramírez B, Becerril S, Salvador J, Portincasa P, Colina I, Gómez-Ambrosi J. Involvement of the leptin-adiponectin axis in inflammation and oxidative stress in the metabolic syndrome.Sci Rep. 2017; 7:6619. doi: 10.1038/s41598-017-06997-0CrossrefMedlineGoogle Scholar
  • 46. Matsui S, Yamane T, Kobayashi-Hattori K, Oishi Y. Ultraviolet B irradiation reduces the expression of adiponectin in ovarial adipose tissues through endocrine actions of calcitonin gene-related peptide-induced serum amyloid A.PLoS One. 2014; 9:e98040. doi: 10.1371/journal.pone.0098040CrossrefMedlineGoogle Scholar
  • 47. Yoshida S, Fuster JJ, Walsh K. Adiponectin attenuates abdominal aortic aneurysm formation in hyperlipidemic mice.Atherosclerosis. 2014; 235:339–346. doi: 10.1016/j.atherosclerosis.2014.05.923CrossrefMedlineGoogle Scholar