The Complement C3a–C3ar Axis Promotes Development of Thoracic Aortic Dissection Via Regulation of MMP2 Expression

The Complement C3a−C3aR Axis Promotes Development of Thoracic Aortic Dissection via Regulation of MMP2 Expression

This information is current as Weihong Ren, Yan Liu, Xuerui Wang, Chunmei Piao, of September 24, 2021. Youcai Ma, Shulan Qiu, Lixin Jia, Boya Chen, Yuan Wang, Wenjian Jiang, Shuai Zheng, Chang Liu, Nan Dai, Feng Lan, Hongjia Zhang, Wen-chao Song and Jie Du J Immunol 2018; 200:1829-1838; Prepublished online 24

January 2018; Downloaded from doi: 10.4049/jimmunol.1601386 http://www.jimmunol.org/content/200/5/1829

Supplementary http://www.jimmunol.org/content/suppl/2018/01/24/jimmunol.160138 http://www.jimmunol.org/ Material 6.DCSupplemental References This article cites 54 articles, 21 of which you can access for free at: http://www.jimmunol.org/content/200/5/1829.full#ref-list-1

Why The JI? Submit online. by guest on September 24, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2018 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The Complement C3a–C3aR Axis Promotes Development of Thoracic Aortic Dissection via Regulation of MMP2 Expression

Weihong Ren,*,†,1 Yan Liu,*,1 Xuerui Wang,‡,1 Chunmei Piao,* Youcai Ma,* Shulan Qiu,* Lixin Jia,* Boya Chen,* Yuan Wang,* Wenjian Jiang,* Shuai Zheng,* Chang Liu,* Nan Dai,* Feng Lan,* Hongjia Zhang,* Wen-chao Song,x and Jie Du*

Thoracic aortic dissection (TAD), once ruptured, is devastating to patients, and no effective pharmaceutical therapy is available. Anaphylatoxins released by complement activation are involved in a variety of diseases. However, the role of the complement Downloaded from system in TAD is unknown. We found that plasma levels of C3a, C4a, and C5a were significantly increased in patients with TAD. Elevated circulating C3a levels were also detected in the developmental process of mouse TAD, which was induced by b-aminopropionitrile monofumarate (BAPN) treatment, with enhanced expression of C1q and properdin in mouse dissected aortas. These findings indicated activation of classical and alternative complement pathways. Further, expression of C3aR was obviously increased in smooth muscle cells of human and mouse dissected aortas, and knockout of C3aR notably inhibited BAPN- induced formation and rupture of TAD in mice. C3aR antagonist administered pre- and post-BAPN treatment attenuated the http://www.jimmunol.org/ development of TAD. We found that C3aR knockout decreased matrix metalloproteinase 2 (MMP2) expression in BAPN-treated mice. Additionally, recombinant C3a stimulation enhanced MMP2 expression and activation in smooth muscle cells that were subjected to mechanical stretch. Finally, we generated MMP2-knockdown mice by in vivo MMP2 short hairpin RNA delivery using recombinant adeno-associated virus and found that MMP2 deficiency significantly reduced the formation of TAD. Therefore, our study suggests that the C3a–C3aR axis contributes to the development of TAD via regulation of MMP2 expression. Targeting the C3a–C3aR axis may represent a strategy for inhibiting the formation of TAD. The Journal of Immunology, 2018, 200: 1829–1838.

horacic aortic dissection (TAD) is the most common acute space of the thoracic aorta, leading to separation of the layers thoracic aortic syndrome, with high morbidity and mor- within the aortic wall, which characterizes TAD. Microscopically, by guest on September 24, 2021 T tality. At the macroscopic level, blood enters the medial the pathology of TAD shows medial degeneration in the form of smooth muscle cell (SMC) loss and elastic fiber fragmentation or depletion (1–3). Clinically, the risk factors for TAD are divided *Beijing Anzhen Hospital, Capital Medical University, Beijing Institute of Heart Lung and Blood Vessel Diseases, The Key Laboratory of Remodeling-Related Car- into two categories: conditions that increase aortic wall stress and diovascular Diseases, Ministry of Education, Beijing Collaborative Innovative Re- those that lead to medial degeneration (4). Medial changes, which † search Center for Cardiovascular Diseases, Beijing 100029, China; Department of can result from excessive protease-mediated destruction of the Pathology, Aviation General Hospital of China Medical University, Beijing 100012, China; ‡Beijing Hospital of Traditional Chinese Medicine, Capital Medical Univer- extracellular matrix (ECM), weaken the aortic wall and make it sity, Beijing 100010, China; and xDepartment of Pharmacology and Institute for prone to rupture. Proteolytic enzymes, such as matrix metal- Translational Medicine and Therapeutics, Perelman School of Medicine, University loproteinases (MMPs), can primarily degrade various compo- of Pennsylvania, Philadelphia, PA 19104 nents of the ECM (5) and are implicated in the pathogenesis of 1W.R., Y.L., and X.W. contributed equally to this work. aortic aneurysm and dissection (6–10). MMP9, which is pre- ORCIDs: 0000-0001-8938-3420 (X.W.); 0000-0002-2975-7336 (C.P.); 0000-0003- 3421-3226 (Y.M.); 0000-0002-9143-4909 (S.Z.); 0000-0002-9038-4014 (F.L.). dominantly released from macrophages and neutrophils, has been well studied in the process of TAD and aneurysm. Genetic Received for publication August 9, 2016. Accepted for publication December 28, 2017. and pharmaceutical depletion of MMP9 can protect against the b This work was supported by the Beijing Nova Program (Grant Z151100000315067), effects of -aminopropionitrile monofumarate (BAPN), together the National Natural Science Foundation of China (Grants 81770250, 91339000, with angiotensin II–induced TAD (9). MMP2, which is mainly 81400194, and 81770468), the Chinese Ministry of Science and Technology (Grant from SMCs, increases immediately after the onset of aortic 2016YFC0903000), the Beijing Natural Science Foundation (Grants 7142030 and 7162030), and the Program for Changjiang Scholars and Innovative Research Team dissection (11). Using neutralizing Ab against MMP2 or in University (Grant IRI1074). knockout of MMP2 can prevent aneurysms in Marfan mice Address correspondence and reprint requests to Dr. Jie Du, Beijing Institute of Heart (12, 13). Therefore, rigorous regulation of MMP expression and Lung and Blood Vessel Diseases, Beijing Anzhen Hospital, 2 Anzhen Road, activity is crucial for homeostasis of the ECM. MMP expression Chaoyang District, Beijing 100029, China. E-mail address: [email protected] can be regulated by multiple factors, including proinflammatory The online version of this article contains supplemental material. cytokines, growth factors, and hypoxia (4, 14). Abbreviations used in this article: AAA, abdominal aortic aneurysm; AP, alternative pathway; BAPN, b-aminopropionitrile monofumarate; CP, classical pathway; ECM, Infiltration of various immune cells has been found in the media extracellular matrix; KO, knockout; LP, lectin pathway; MMP, matrix metalloprotei- and adventitia of dissected aortic specimens and may contribute to nase; MRI, magnetic resonance imaging; qPCR, quantitative PCR; SMC, smooth aortic rupture (1). The complement system is an important driver muscle cell; TAD, thoracic aortic dissection; WT, wild-type. of inflammation and is capable of eliciting a proinflammatory Copyright Ó 2018 by The American Association of Immunologists, Inc. 0022-1767/18/$35.00 response by recruitment of immune cells. Complement is activated www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601386 1830 KNOCKOUT OF C3aR PREVENTS DEVELOPMENT OF TAD through three pathways, including the classical pathway (CP), the Magnetic resonance imaging (MRI) was performed using an M2 compact lectin pathway (LP), and the alternative pathway (AP). All three MRI system (Aspect Imaging, Shoham, Israel) with mice in the prone pathways converge at a central enzyme, C3 convertase, which position on top of a microscopy single-loop surface coil (diameter = 35 mm). Morphology of the thoracic aorta was assessed by imaging, as previously enzymatically cleaves C3 (15, 16). Among the three pathways, AP described (24). is involved in elastase-induced development of abdominal aortic Measurement of plasma anaphylatoxin levels aneurysm (AAA), in which the AP C3 convertase components factor B and properdin play a critical role (17, 18). The anaphy- Plasma levels of anaphylatoxin C3a, C4a, and C5a in patients with TAD and latoxin C3a, which is released by C3 cleavage, is regarded as a healthy controls were determined using a BD Cytometric Bead Array proinflammatory stimulator in a variety of diseases (19, 20). (BBA) Human Anaphylatoxin Kit (BD Biosciences, Franklin Lakes, NJ), according to the instruction manual. Plasma from BAPN- or vehicle-treated Blockade of C3a and C5a activity together protects against de- mice was collected, and C3a levels were determined by ELISA (Uscn, velopment of AAA in mice, but inhibition of C3a or C5a alone has Wuhan, China), as previously described (25, 26). no effect (17). The pathogenesis of TAD is not completely the Histopathology same as that of AAA, and whether anaphylatoxins released by complement activation are involved in the development of TAD Mouse aortas were fixed in 10% formalin, embedded in paraffin, and m has not been studied. sectioned at 5 m thickness. Immunohistochemical staining was performed as previously described (21, 25). In brief, sections were treated In this study, we aimed to determine the role of the C3a–C3aR with xylene to remove the paraffin and were rehydrated, incubated with 3% axis in the formation and rupture of TAD. We found that the H2O2 for 10 min at room temperature, and washed three times with PBS. C3a–C3aR axis was activated in the mouse model and human Then sections were blocked with serum for 30 min and incubated with a TAD specimens. Genetic deletion of C3aR attenuated medial primary Abs against SMA (1:200 dilution; ZSGB-Bio, Beijing, China) Downloaded from and Mac3 (Santa Cruz Biotechnology, Dallas, TX), followed by incubation degeneration and prevented rupture of TAD via downregulation of with AEC staining solution (ZSGB-Bio). Images were captured and further MMP2. Additionally, pharmacological inhibition of C3aR in the analyzed using Image-Pro Plus 3.0 (ECIPSE80i/90i). mouse model had a similar effect. Our data suggest that targeting For cryostat sections, mouse aorta samples were fixed in 4% parafor- the C3a–C3aR axis may represent a strategy for inhibiting TAD maldehyde, embedded in optimum cutting temperature compound, frozen in 2 formation. liquid nitrogen, and stored at 80˚C until sectioning. Immunofluorescence costaining of C3aR (1:200 dilution; Abcam, Cambridge, MA) with aSMA (1:200 dilution; Sigma-Aldrich), neutrophil (1:200 dilution; Abcam), or http://www.jimmunol.org/ Materials and Methods CD31 (1:200 dilution; BD Biosciences) on sections of the aorta was performed and viewed with a confocal fluorescence microscope (Leica Specimens from patients Microsystems, Buffalo Grove, IL). Blood samples were collected from newly diagnosed patients with TAD prior to rupture and before treatment intervention was initiated, as well as Elastin staining from age-matched non-TAD individuals undergoing regular physical ex- Elastin fragmentation was graded based on the degree of elastin filament aminations at Beijing Anzhen Hospital. Aortic dissection samples were breakage, as described previously (21). Elastin in the thoracic aorta was collected before rupture from patients with TAD undergoing repair surgery stained using Gomori’s aldehyde-fuchsin staining method and an elastic at Beijing Anzhen Hospital. None of the patients had a known genetic fiber staining kit (Fuzhou Maixin Biotech, Fuzhou, China). Briefly, after syndrome related to aortic disease, such as Marfan, Turner, Loeys–Dietz, or deparaffinization and rehydration, sections were incubated for 5 min in by guest on September 24, 2021 Ehlers–Danlos syndrome. The control aortas were trimmings that were Lugol’s iodine solution, washed with PBS, and incubated with sodium discarded during heart-transplantation surgeries. The basic demographics thiosulfate for 5 min. After washing with PBS and 70% ethanol, the sec- of the patients and controls are shown in Supplemental Table I. Informed tions were incubated with aldehyde–fuchsin for 10 min and acid Orange G consent was obtained for use of these specimens. Use of human blood and for 5 s. tissue was approved by the Medical Ethical Committee of Capital Medical University and was in compliance with the principles outlined in the SMC culture and treatment Declaration of Helsinki (21). SMCs were isolated from C57BL/6 WT mice, as previously described (27). Animal model Briefly, mice were sacrificed with CO2 narcosis, and aortas were excised and then digested with type II collagenase at 37˚C for 30 min to remove C3aR-knockout (KO) mice on a C57BL/6 background were generated as the adventitia. The endothelium was removed by gently rubbing the intima previously described (22). The experiments were approved by the Insti- with a sterile cotton-tipped applicator, followed by further digestion with a tutional Animal Care and Use Committee of Capital Medical University mixture of collagenase and elastase for 30 min. Cells were then cultured and conformed to the National Institutes of Health’s Guide for the Care in DMEM (Life Technologies, Carlsbad, CA) supplemented with 10% and Use of Laboratory Animals. Mice were maintained under a controlled inactivated FBS and 1% penicillin-streptomycin. For cyclic stretching of temperature and a 12-h light/dark cycle, with free access to water and SMCs, cells were cultured on silicone elastomer-bottom collagen-coated standard laboratory chow. Three-week-old male C3aR-KO mice and their plates (Flexcell, Hillsborough, NC) at 37˚C overnight. Then cells were wild-type (WT) littermates were administered BAPN (Sigma-Aldrich, St. subjected to cyclic mechanical stretching using a computer-controlled Louis, MO) in the drinking water, at a dose of 1 g/kg/d for 4 wk, to induce mechanical strain unit (Flexcell 5000) at 18% elongation for 12 h, as aortic dissection (9), with modifications. For C3aR antagonist treatment, previously described (21). SMCs were then treated with recombinant SB290157 (Sigma-Aldrich) dissolved in 50% DMSO was administered to murine C3a (10 ng/ml) and cultured for 24 h under stretch. male C57BL/6 mice (1 mg/kg/d) by i.p. injection every other day from 3 d For MMP2 knockdown, mouse SMCs were transfected with mouse MMP2 before or 1 wk after BAPN administration to the end of the experiments small interfering RNA (siRNA) (sense: 59-CAUACAGGAUCAUUGGUUATT- (23). For MMP2 silencing, 1011 genome copies of recombinant adeno- 39, anti-sense: 59-UAACCAAUGAUCCUGUAUGTG-39)orscramblesiRNA associated virus containing MMP2 short hairpin RNA (shRNA) or using a Lonza 4D-Nucleofector X-unit system (Basel, Switzerland), according scramble shRNA was administered to 19-d-old C57BL/6 mice by retro- to the manufacturer’s instructions (28). Briefly, a 1 3 106 cell pellet was orbital injection. Then mice were treated with BAPN for 4 wk. At the end resuspended in 100 mlofP1primarycellbufferandthen30pmolsiRNAwas of the experiments, mice were sacrificed using CO2, and blood and aortas added to the cell suspension. The mixture was transferred to the reaction cu- were collected. Blood clots in the aortic wall were regarded as formation vette, and electroporation was carried out. After completion of the run, cells of TAD. were seeded into one well of a six-well plate. Thirty-six hours after electroporation, mRNA was harvested and knockdown efficiency was determined by Aortic ultrasonography and magnetic resonance imaging quantitative PCR (qPCR). Mice were anesthetized with 1% isoflurane and underwent echography in qPCR the M-mode, using a high-resolution microultrasound system (Vevo 2100; VisualSonics, Toronto, ON, Canada) equipped with a 30-MHz transducer. Total RNA was extracted from thoracic aorta samples of human patients, A color Doppler examination was performed to detect arterial flow, as experimental mice, or SMCs using TRIzol Reagent (Life technologies). previously described (21). A total of 2 mg of RNA was reversed transcribed using a GoScript Reverse The Journal of Immunology 1831

Transcription System (Promega, Madison, WI), according to the manu- wk of BAPN treatment compared with control aortas (Fig. 1F) facturer’s instructions. An iQ5 system (Bio-Rad, Hercules, CA) with This finding suggested activation of the CP and AP of the com- SYBR Green I (Takara, Shiga, Japan) was used for qPCR. Amplification plement system in TAD mice. was performed using 40 cycles of 95˚C for 5 min, 95˚C for 45 s, and 60˚C for 1 min. The housekeeping gene GAPDH was used as a control. The C3aR expression in SMCs is increased in TAD primers used are shown in Supplemental Table II. C3a functions by binding to its receptor C3aR, and enhanced Western blot analysis expression of C3aR in diseased tissues may amplify its downstream Mouse and human aorta or cultured SMCs were harvested, snap-frozen in biological effect (15). We found that, in human TAD samples, liquid nitrogen, and stored at 280˚C. Protein was extracted using a protein mRNA (Fig. 2A) and protein (Fig. 2B) C3aR levels were higher in extraction kit containing Protease and Phosphatase Inhibitor Cocktail TAD patients than in controls. This finding suggested that en- (Pierce Biotechnology, Rockford, IL). Equal amounts of protein extract were separated using a 10% SDS–PAGE gel. The blots were probed with hanced expression of C3aR, together with increased C3a levels, the primary Abs anti-GAPDH (1:2000 dilution; Sigma-Aldrich), anti- may contribute to the development of TAD. We next examined C3aR (1:1000 dilution; Santa Cruz Biotechnology), and anti-MMP2 C3aR expression in the aortas of mice with BAPN administration (1:1000 dilution; Santa Cruz Biotechnology) and then probed with IR for 1, 2, and 4 wk and found that protein C3aR levels were in- dye–conjugated secondary Abs (1:5000 dilution; Rockland Immuno- chemicals, Gilbertsville, PA) for 1 h. The blots were washed, exposed, and creased after 4 wk of BAPN treatment (Fig. 2C). Immunostaining analyzed using an Odyssey infrared imaging system (LI-COR Biosciences, of C3aR verified increased C3aR protein expression in the media Lincoln, NE). of aortas from mice treated with BAPN for 4 wk (Fig. 2D). Fur- thermore, we observed strong C3aR protein expression in non- Statistical analysis ruptured and ruptured TAD (Fig. 2E). Downloaded from All data are shown as mean 6 SEM. Comparisons between two groups C3aR is expressed in myeloid cells and nonmyeloid cells (20, 22, 35). were analyzed with the Student t test, and one-way ANOVA was used to x2 To identify the cellular location of C3aR in TAD, we costained compare multiple groups. The test was used to analyze the incidence of a TAD and rupture. Values of p , 0.05 were regarded as significant. C3aR with CD31 (endothelial cell marker), neutrophil, or SMA (SMC marker) in aortic slices from mice treated with BAPN for 4 wk. We observed C3aR expression in SMCs and neutrophils, but Results not in endothelial cells (Fig. 3). Because low C3aR levels were http://www.jimmunol.org/ Complement is activated in human and mouse TAD detected in SMCs from control aortas (Fig. 2D), we concluded that Previous studies have shown the presence of complement proteins C3, C3aR expression was induced in SMCs during the development of C4, factor B, properdin, and C5-9 neoantigen in the aortic wall of TAD. A previous study showed that few neutrophils had infiltrated human AAAs, indicating activation of the CP and AP (17, 18, 29–31). nondissected aortas from mice treated with BAPN (9). We also To evaluate activation of the complement system in patients with observed minimal infiltration of neutrophils in control and non- TAD, we measured circulating levels of the anaphylatoxins C3a, dissected aortas (Supplemental Fig. 1). The neutrophils are found C4a, and C5a in patients with nonruptured TAD and healthy control mainly in the media of dissected aorta (dissected area) and could volunteers. We found that the levels of these anaphylatoxins were also be observed in adventitia of dissected aorta. These data significantly elevated in patients with TAD, which suggested ex- suggested that the C3a–C3aR axis in SMCs played a vital role in by guest on September 24, 2021 cessive and ongoing complement activation (Fig. 1A). the development of TAD. Based on our findings in human patients, we sought to validate our results in a mouse model of TAD. We administered BAPN, a C3aR deficiency or inhibition protects against development lysyl oxidase inhibitor, to 3-wk-old mice for 4 wk to induce TAD and rupture of TAD (9, 32). BAPN treatment prevents formation of elastin and colla- Because we found increased production of C3a and expression of gen cross-linking by inhibiting the activity of lysyl oxidase (33, C3aR in TAD mouse SMCs, we wished to determine the role of the 34). Although no single animal model can completely mimic C3a–C3aR axis in the pathogenesis of TAD. We fed WT and C3aR- human TAD, the BAPN-induced experimental TAD model has KO mice BAPN for 4 wk. We observed that the incidence of TAD many features of this disease (9, 32). Using color Doppler ultra- was significantly decreased from 87.5 to 17% by C3aR KO, and no sonography, we observed disrupted aortic walls and disordered sudden death caused by aortic rupture occurred in C3aR-KO mice flow direction in the ascending aorta and aortic arch of BAPN- (Fig. 4A). This finding suggested that genetic deletion of C3aR had treated mice (Fig. 1B). A larger aortic diameter was also visible on a remarkable effect on inhibiting the development of BAPN- MRI (Fig. 1C). We isolated aortas from BAPN-treated mice and induced TAD. Corresponding with the above-mentioned data, found that they were dilated from ascending to thoracic portions, elastin staining showed less fragmentation in BAPN-treated C3aR- with thrombosis in the vessel wall (Fig. 1D). These results sug- KO mice compared with BAPN-induced TAD in WT mice gested that the TAD mouse model was successfully established. (Fig. 4B). Similarly, immunohistochemical staining of aSMA We measured plasma C3a levels in mice at 1, 2, and 4 wk after revealed a lower extent of SMC depletion in BAPN-treated KO BAPN treatment. Plasma C3a levels were elevated as early as 1 wk mice than in similarly treated WT mice (Fig. 4C). Taken together, after BAPN administration and were maintained at similar levels at these results indicated that C3aR KO could dramatically protect the end of 2 wk. A robust increase in C3a levels was observed at the against medial degeneration and prevent the development of TAD. end of 4 wk with BAPN treatment, when most mice had formed We also stained macrophages with Ab against Mac3 on aortic TAD (Fig. 1E). To further evaluate activation of the complement sections from WT and C3aR-KO mice that were treated with BAPN system in the formation and rupture of TAD, we next examined for 4 wk. We found that C3aR KO significantly attenuated the in- circulating C3a levels in mice with rupture of TAD. Even higher filtration of macrophages (Fig. 4F), which might inhibit the con- plasma C3a levels were detected in mice with rupture of TAD tribution of inflammation to the development of aortic dissection. compared with those with nonruptured TAD (Fig. 1E). These re- Additionally, because C5a was significantly elevated in the plasma sults indicated activation of the complement system that was ac- of patients with TAD, the contribution of C5a/C5aR to the forma- companied by formation of BAPN-induced TAD, which was tion of TAD could not be ignored. We observed that C5aR KO enhanced in the process of rupture. We also found increased C1q partially reduced the formation of TAD and rupture (data not and properdin mRNA expression in the aortas of TAD mice with 4 shown; the subject is being pursued separately). 1832 KNOCKOUT OF C3aR PREVENTS DEVELOPMENT OF TAD Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 1. The complement system is activated in patients with TAD and in BAPN-treated mice. (A) Plasma C3a, C4a, and C5a levels in patients with

TAD and healthy controls (Ctrl) were assayed using a BD Cytometric Bead Array (BBA) Human Anaphylatoxin Kit. Data represent the mean 6 SEM; nCtrl =7,nTAD = 12. C57BL/6 mice were administered BAPN for 4 wk. (B) The direction of ﬂow in the aortas of these mice was detected by color Doppler ultrasonography (white lines indicate the aortic walls, and the green arrowhead denotes an area without ﬂow). (C) MRI images of aortas (white lines indicate aortic walls). (D) Representative macroscopic images show blood clot formation in a dissected thoracic aorta induced by 4 wk of BAPN treatment. (E) Plasma levels of the anaphylatoxin C3a in mice treated with BAPN for 1 wk (BAPN 1w), 2 wk (BAPN 2w), and 4 wk (BAPN 4w) were assayed by

ELISA. Data represent the mean 6 SEM; nCtrl = 10, nBAPN 1w = 10, nBAPN 2w =7, nBAPN 4w = 13, nNonrupture =9,nRupture =7.(F) C1q and properdin mRNA expression in aortas was assessed by qPCR and expressed as fold induction compared with controls after being normalized to GAPDH. Data represent the mean 6 SEM; n =6.*p , 0.05, **p , 0.01 versus Ctrl. ##p , 0.01 versus nonrupture.

We next examined whether pharmacological inhibition of BAPN-treated WT mice. Pretreatment with SB290157 markedly C3aR could play a similar role as C3aR KO. We administered attenuated the incidence of TAD from 87 to 42%, and aortic rupture SB290157, an inhibitor of C3aR (23), at different time intervals to decreased from 37 to 25%, although this was not signiﬁcant The Journal of Immunology 1833 Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 2. C3aR expression is increased in aortas in patients with TAD and in BAPN-treated mice. (A) C3aR mRNA levels in human dissected tissues were assessed by qPCR and expressed as fold induction compared with controls after being normalized to GAPDH. Data represent the mean 6 SEM; n =5. (B) C3aR protein expression levels in human dissected aortas were detected and presented as fold induction compared with controls after being normalized to GAPDH. Data represent the mean 6 SEM; n =5.(C) C3aR protein expression levels in aortas of mice treated with BAPN for different time points were assessed and presented as fold induction compared with controls after being normalized to GAPDH. Data represent the mean 6 SEM; n = 6. C3aR expression in aortas from BAPN-treated mice (D) and in ruptured and nonruptured aortas (E) was examined with a speciﬁc Ab and an Alexa Fluor 555– labeled secondary Ab. Nuclei were stained with DAPI; n = 6. Scale bars, 100 mm. *p , 0.05, **p , 0.01 versus control (Ctrl). L, aortic lumen.

(Fig. 4D). When we administered SB290157 from 1 wk post-BAPN (Fig. 5A). Similarly, in the BAPN-induced mouse TAD model, we treatment until the end of the experiment, the incidence of TAD was found that pro-MMP2 levels were elevated as early as 1 wk after still markedly reduced (from 90 to 50%, Fig. 4E), whereas the BAPN treatment, whereas the activated form of MMP2 was ele- difference in the rate of rupture was not signiﬁcantly changed. vated after 4 wk of BAPN administration compared with controls These results indicated that the C3a–C3aR axis played a detrimental (Fig. 5B). We also determined MMP2 expression levels in aortas role in the development of TAD. of C3aR-KO mice after 4 wk of BAPN treatment and found that they were markedly decreased compared with the aortas of WT C3aR affects TAD through regulation of MMP2 mice (Fig. 5C). These results indicated that the C3a–C3aR axis MMPs play a vital role in the pathogenesis of aortic dissection, contributed to aortic dissection by regulating MMP2 expression. because MMP9 KO or inhibition notably decreases the incidence of To determine whether the C3a–C3aR axis regulated MMP2 acute aortic dissection (9). To evaluate whether MMPs are in- expression directly in SMCs, we stimulated SMCs with volved in the effect of the C3a–C3aR axis on the development and recombinant C3a. We found that C3a alone did not elevate MMP2 rupture of TAD, we measured MMP2 and MMP9 expression expression (data not shown). During the developmental process of levels in aortic tissues of patients with TAD and found that they TAD, the wall of the aorta is weakened by degradation of ECM, were elevated in TAD tissues compared with control aortas and SMCs are subjected to increased mechanical stretch. Therefore, 1834 KNOCKOUT OF C3aR PREVENTS DEVELOPMENT OF TAD Downloaded from

FIGURE 3. Increased C3aR protein expression in SMCs of aortas from BAPN-treated mice. Cellular localization of C3aR in aortas of mice treated with BAPN for 4 wk was detected by costaining of C3aR with CD31, aSMA, or neutrophil with primary Abs, followed by Alexa Fluor 488–labeled or Alexa http://www.jimmunol.org/ Fluor 555–labeled secondary Ab. Nuclei were stained with DAPI; n = 6. Scale bars, 100 mm. L, aortic lumen. we treated SMCs with C3a following mechanical stretch (18% AAA formation (17). This suggests that the AP of complement elongation for 12 h). Interestingly, we found that C3a treatment of plays a vital role in AAA (17). The AP C3 convertase is formed by SMCs under mechanical stretch induced MMP2 expression. This three components: factor B, C3b, and properdin. A recent study result is consistent with the in vivo data, which showed a regulatory showed that properdin was required for the development of AAA effect of the C3a–C3aR axis on MMP2 in the development and as a C3 convertase stabilizer (18). Another study demonstrated rupture of TAD (Fig. 5D). that natural IgG binding to fibrinogen located in elastase-perfused by guest on September 24, 2021 SMC-derived MMP2 plays a major role in angiotensin II– or aortic tissues activated the complement LP, with increases in C1q nicotine-induced mouse AAA (36); however, the effect of MMP2 and MBL to induce AAA (37). This finding implicates the CP and on the development of TAD remains unknown. Therefore, we LP in the development of AAA. Furthermore, C2, a component of generated MMP2-deficient mice by in vivo shRNA delivery. We the LP and CP, is increased in human AAA tissues (29). Although first determined the knockdown effect of MMP2 siRNA in cul- all three pathways of complement are known to be involved in the tured mouse SMCs (Fig. 6A). We then compared adeno-associated development of AAA, less is known regarding the potential role of virus containing MMP2 shRNA with knockdown MMP2 in vivo. complement in aortic dissection. A previous study suggested that MMP2 shRNA significantly attenuated MMP2 mRNA expression C4d deposition (complement activation) is a predictive marker for in mouse aortas (Fig. 6B). Moreover, MMP2 KO markedly re- susceptibility to ascending aortic dissection (38). duced the formation of BAPN-induced TAD, and rupture was also All three pathways of complement activation lead to the for- attenuated from 42 to 8.3% (Fig. 6C, 6D). Accordingly, MMP2 mation of C3 convertase and release of C3a and C3b. C3a, C4a, and shRNA inhibited aortic elastin degradation (Fig. 6E). These data C5a are recognized as anaphylatoxins and are potent inflammatory indicated the vital role of MMP2 in BAPN-induced TAD and mediators. Increased levels of C3, as well as C3a, have been ob- provided evidence for the C3a–C3aR axis in mediating the de- served in human AAA thrombus, suggesting deposition and acti- velopment of TAD by regulating MMP2 in SMCs. vation of C3 in AAA (17). We also found elevated levels of C3a in the plasma of patients with TAD and in BAPN-induced TAD mice. Discussion The latter results are consistent with the observation in AAA and In this study, we showed that the complement system was activated indicate that complement C3a may participate in the pathogenesis in patients with TAD and in a BAPN-induced mouse TAD model of TAD. C3a binds to its cognate receptor C3aR to trigger with elevated release of C3a. Moreover, expression of C3aR was downstream biological effects. C3aR is expressed in multiple or- increased in SMCs of TAD mice, and KO or inhibition of C3aR gans and cells, including endothelial cells, SMCs, and leukocytes, reduced the development of TAD. Finally, the C3a–C3aR axis under distinct conditions (39). In our study, there was low ex- mediated MMP2 expression in SMCs and TAD aortas, and pression of C3aR in SMCs from normal aortas; however, it was knockdown of MMP2 also attenuated the formation of TAD. Our dramatically increased in aortic dissection, which suggested that results showed that the C3a–C3aR axis played a detrimental role the C3a–C3aR axis and promotion of TAD mainly occurred in in TAD. SMCs. We also observed C3aR expression in neutrophils; Multiple studies have indicated that the complement cascade however, previous studies on the function of C3a in neutrophils participates in the pathogenesis of AAA (17, 18, 30, 37). Defi- are complicated. C3a has proinflammatory effects in mast cells, ciency of factor B (an essential component of the AP), but not C4 eosinophils, monocytes/macrophages, and APCs by inducing (a component of the LP and CP), protects against elastase-induced the release of inflammatory cytokines or degranulation (40–42); The Journal of Immunology 1835 Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 4. KO or pharmacological inhibition of C3aR protects against development of BAPN-induced TAD. Three-week-old WT and C3aR-KO mice were administered BAPN or vehicle in the drinking water for 4 wk. (A) Representative macroscopic views showing that depletion of C3aR prevents the 2/2 2/2 formation and rupture of BAPN-induced TAD. nWT Ctrl =6,nC3aR Ctrl =6,nWT BAPN = 16, nC3aR BAPN = 12. **p , 0.01 versus WT BAPN for the # incidence of TAD, p , 0.05 versus WT BAPN for rupture. (B) Fragmentation of elastin in thoracic aortas was detected by aldehyde-fuchsin staining; nWT =6, 2/2 2/2 nC3aR = 7. Scale bars, 100 mm. **p , 0.01 versus WT. (C) SMC loss grade in thoracic aortas was evaluated by staining of aSMA; nWT =7,nC3aR =8. Scale bars, 100 mm. **p , 0.01 versus WT. (D) C57BL/6 mice were treated with SB290157, a C3aR antagonist, 3 d before administration of BAPN by i.p. injection every other day until the end of the experiments. Aortas of mice were dissected at the end of the experiments, and the rates of TAD formation and rupture were calculated; nCtrl = 10, nBAPN = 20, npretreatment = 12. *p , 0.05 versus WT BAPN for the incidence of TAD. (E) C57BL/6 mice were treated with SB290157 1 wk post-BAPN administration. Aortas of mice were dissected at the end of the experiments, and the rates of TAD formation and rupture were + calculated; nCtrl =5,nBAPN = 10, nposttreatment = 10. *p , 0.05 versus WT BAPN for the incidence of TAD. (F) Mac3 areas in aortas from C3aR-KO mice treated with BAPN for 4 wk were detected and presented as fold induction compared with BAPN-treated WT mice; n = 6. Scale bars, 100 mm. **p , 0.01 versus WT. however, it also has an anti-inﬂammatory effect in neutrophils by Plasma C3a levels were elevated as early as after 1 wk of BAPN attenuating their mobilization into the circulation following injury treatment. Although C3aR expression in the aorta did not increase (43), and C3a cannot chemoattract or stimulate degranulation of after 2 wk of BAPN treatment, basal C3aR expression was observed. neutrophils (40, 44). Our data and a previous study (9) showed that Therefore, increased C3a expression might exert its effects through few neutrophils inﬁltrated into aortas before the formation of basal expression of C3aR in the early stage of TAD formation. Plasma dissection in BAPN-treated mice, which suggested that C3a is less C3a levels and aortic C3aR expression were markedly elevated during important in neutrophils than in SMCs in the development of the late stage, which suggested that the C3a–C3aR axis could con- TAD. tribute to dissection. In our study, C3aR deletion and C3aR antagonist pretreatment A major pathological feature of TAD is medial degeneration, blocked the C3a–C3aR axis before aortic injury, as well as the in- which is characterized by SMC loss and ECM degradation. Ac- crease in C3a and C3aR expression initiated by BAPN administration. cordingly, the function of SMCs is critical for the development of 1836 KNOCKOUT OF C3aR PREVENTS DEVELOPMENT OF TAD Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 5. C3a regulates MMP2 expression in aortas of BAPN-treated mice and in SMCs. (A) MMP2 and MMP9 mRNA levels in human dissected tissues were assessed with qPCR and expressed as fold induction compared with controls after being normalized to GAPDH. Data represent the mean 6 SEM; n =5.(B) Protein expression of MMP2 in aortas from BAPN-treated mice was detected and presented as fold induction compared with controls after being normalized to GAPDH; n =6.(C) Protein expression of activated MMP2 in aortas from C3aR-KO mice treated with BAPN for 4 wk was detected and presented as fold induction compared with WT mice after being normalized to GAPDH; n =6.(D) Immunohistochemical staining of MMP2 was performed on aortic slices of BAPN-treated mice. MMP2+ areas in aortas from C3aR-KO mice treated with BAPN for 4 wk were detected and presented as fold induction compared with BAPN-treated WT mice; n =6.**p , 0.01 versus WT. Scale bars, 100 mm. (E) SMCs that were subjected to mechanical stretch (18% elongation) were stimulated with recombinant C3a for 24 h. Protein levels of activated MMP2 were detected and presented as fold induction compared with controls after being normalized to GAPDH. Data represent the mean 6 SEM; three independent experiments were performed. *p , 0.05 versus stretch.

TAD (3). SMCs switch from a contractile phenotype to a synthetic also found that BAPN treatment increased MMP9, as well as phenotype, which secretes various substances, including ECM MMP2, expression in the aorta. Although C3aR KO attenuated proteins, growth factors, inflammatory cytokines, and proteases MMP9 mRNA expression in aortas (Supplemental Fig. 2), infil- (45, 46). Excessive proteases, such as MMPs, mediate destruction tration of neutrophils and macrophages was also decreased. of the ECM, and could weaken the aortic wall (3). Various studies Neutrophils and macrophages are major sources of MMP9; have shown increased expression of MMPs in TAD, including therefore, the decrease in MMP9 mRNA expression in C3aR-KO MMPs 1, 2, 9, and 12 (6, 7, 47–49). aortas could have resulted from reduced infiltration of macro- Kurihara et al. (9) showed an important role for MMP9 in the phages and neutrophils. SMCs are a major source of MMP2 in development of TAD. They found that TAD was closely associ- vascular tissues (36). Our study also showed increased MMP2 ated with increased activity of MMP9 in aortas of mice, and KO of expression in mouse TAD tissues, and KO of C3aR decreased MMP9 protected against the formation and rupture of TAD. We MMP2 mRNA expression in aortas of TAD mice. Therefore, The Journal of Immunology 1837 Downloaded from

FIGURE 6. Knockdown of C3aR attenuates BAPN-induced formation of TAD. (A) Primary mouse SMCs were transfected with MMP2 or scramble siRNA for 36 h. MMP2 mRNA expression is shown as the fold change compared with the scramble siRNA group after normalization to GAPDH. Data represent the mean 6 SEM; three independent experiments were performed. **p , 0.01. (B) C57BL/6 mice were injected with 1011 genome copies of http://www.jimmunol.org/ adeno-associated virus (AAV) containing MMP2 shRNA or AAV containing scramble shRNA 3 d before BAPN treatment. At the end of 4 wk of BAPN treatment, MMP2 mRNA expression in aortas was detected and shown as the fold change compared with the AAV scramble shRNA group after normalization to GAPDH; n =4.*p , 0.05. (C) Representative macroscopic view shows that depletion of C3aR prevents the formation of BAPN-induced TAD. nAAV-scramble shRNA Ctrl =4,nAAV-MMP2 shRNA Ctrl =4,nAAV-scramble shRNA BAPN = 12, nAAV-MMP2 shRNA BAPN =12.*p , 0.05 versus AAV scramble shRNA BAPN for the incidence of TAD. (D) Fragmentation of elastin in thoracic aortas was detected by aldehyde-fuchsin staining; n = 6. Scale bars, 100 mm. reduced expression of MMP2 and MMP9 represents the under- Disclosures lying mechanism for the protective effect of C3aR KO on the The authors have no financial conflicts of interest. development of TAD. by guest on September 24, 2021 MMP expression can be regulated by various pathways, in- k References cluding NF- B, MAPKs, and SMAD2/3. C3a is a potent 1. Nienaber, C. A., and R. E. Clough. 2015. Management of acute aortic dissection. proinflammatory mediator. However, whether the C3a–C3aR Lancet 385: 800–811. axis regulates MMP2 expression in TAD through its proin- 2. LeMaire, S. A., and L. Russell. 2011. Epidemiology of thoracic aortic dissection. Nat. Rev. Cardiol. 8: 103–113. flammatory effect remains to be investigated. Our study showed 3. Wu, D., Y. H. Shen, L. Russell, J. S. Coselli, and S. A. LeMaire. 2013. Molecular that C3a induced MMP2 expression in stretch-stimulated, but mechanisms of thoracic aortic dissection. J. Surg. Res. 184: 907–924. not static, SMCs. During the formation of TAD, SMCs are 4. Goldfinger, J. Z., J. L. Halperin, M. L. Marin, A. S. Stewart, K. A. Eagle, and V. Fuster. 2014. Thoracic aortic aneurysm and dissection. J. Am. Coll. Cardiol. subjected to increased mechanical stretch as a result of aortic 64: 1725–1739. wall weakening (21). We (21) and other investigators (50, 51) 5. Ra, H. J., and W. C. Parks. 2007. Control of matrix metalloproteinase catalytic have previously reported that mechanical stretch induces en- activity. Matrix Biol. 26: 587–596. 6. Koullias, G. J., P. Ravichandran, D. P. Korkolis, D. L. Rimm, and doplasmic reticulum stress and inflammation in SMCs. There- J. A. Elefteriades. 2004. Increased tissue microarray matrix metalloproteinase fore, C3a may have increased MMP2 expression in SMCs expression favors proteolysis in thoracic aortic aneurysms and dissections. Ann. Thorac. Surg. 78: 2106–2110, discussion 2110–2111. under certain conditions, such as inflammation. Ignatius et al. 7. Sangiorgi, G., S. Trimarchi, A. Mauriello, P. Righini, E. Bossone, T. Suzuki, (52) showed that C3a, together with IL-1b, could stimulate the V. Rampoldi, and K. A. Eagle. 2006. Plasma levels of metalloproteinases-9 and release of IL-6 and IL-8 in osteoblasts, whereas C3a alone -2 in the acute and subacute phases of type A and type B aortic dissection. J. Cardiovasc. Med. 7: 307–315. could not. Moreover, C3a and C5a increase the LPS-induced 8. Martıń-Alonso, M., A. B. Garcıá-Redondo, D. Guo, E. Camafeita, F. Martıńez, production of IL-6 in mouse peritoneal macrophages (53). C3a A. Alfranca, N. Meńdez-Barbero, A´ . Pollań, C. Sańchez-Camacho, D. T. Denhardt, et al. 2015. Deficiency of MMP17/MT4-MMP proteolytic ac- enhances the IFN-stimulated expression of cell surface markers tivity predisposes to aortic aneurysm in mice. Circ. Res. 117: e13–e26. of monocyte-derived dendritic cells, and this is attributed to an 9. Kurihara, T., R. Shimizu-Hirota, M. Shimoda, T. Adachi, H. Shimizu, increase in C3aR expression by IFNs (54). In our study, we S. J. Weiss, H. Itoh, S. Hori, N. Aikawa, and Y. Okada. 2012. Neutrophil-derived matrix metalloproteinase 9 triggers acute aortic dissection. Circulation 126: observed an increase in C3aR expression in SMCs of TAD. 3070–3080. This finding suggests that the C3a–C3aR axis plays a prominent 10. Zhang, X., Y. H. Shen, and S. A. LeMaire. 2009. Thoracic aortic dissection: are role in SMCs. We were not able to identify the exact mecha- matrix metalloproteinases involved? Vascular 17: 147–157. 11. Manabe, T., K. Imoto, K. Uchida, C. Doi, and Y. Takanashi. 2004. Decreased nism by which C3a upregulated MMP2 expression in stretched tissue inhibitor of metalloproteinase-2/matrix metalloproteinase ratio in the SMCs. A potential mechanism may relate to stretch- or acute phase of aortic dissection. Surg. Today 34: 220–225. inflammation-stimulated C3aR expression. This issue remains to 12. Chung, A. W., H. H. Yang, M. W. Radomski, and C. van Breemen. 2008. Long- term doxycycline is more effective than atenolol to prevent thoracic aortic an- be investigated in future studies. eurysm in marfan syndrome through the inhibition of matrix metalloproteinase-2 In conclusion, our study shows that the C3a–C3aR axis is an and -9. Circ. Res. 102: e73–e85. 13. Xiong, W., T. Meisinger, R. Knispel, J. M. Worth, and B. T. Baxter. 2012. MMP- important pathway in regulating the development of TAD via an 2 regulates Erk1/2 phosphorylation and aortic dilatation in Marfan syndrome. effect on MMP2 expression. Circ. Res. 110: e92–e101. 1838 KNOCKOUT OF C3aR PREVENTS DEVELOPMENT OF TAD

14. Piechota-Polanczyk, A., S. Demyanets, M. Mittlboeck, M. Hofmann, 34. Ma¨ki, J. M., R. Sormunen, S. Lippo, R. Kaarteenaho-Wiik, R. Soininen, and C. M. Domenig, C. Neumayer, J. Wojta, M. Klinger, J. Nanobachvili, and I. Huk. J. Myllyharju. 2005. Lysyl oxidase is essential for normal development and 2015. The influence of simvastatin on NGAL, matrix metalloproteinases and function of the respiratory system and for the integrity of elastic and collagen their tissue inhibitors in human intraluminal thrombus and abdominal aortic fibers in various tissues. Am. J. Pathol. 167: 927–936. aneurysm tissue. Eur. J. Vasc. Endovasc. Surg. 49: 549–555. 35. Mathern, D. R., and P. S. Heeger. 2015. Molecules great and small: the com- 15. Ricklin, D., and J. D. Lambris. 2013. Complement in immune and inflammatory plement system. Clin. J. Am. Soc. Nephrol. 10: 1636–1650. disorders: pathophysiological mechanisms. J. Immunol. 190: 3831–3838. 36. Wang, S., C. Zhang, M. Zhang, B. Liang, H. Zhu, J. Lee, B. Viollet, L. Xia, 16. Dunkelberger, J. R., and W. C. Song. 2010. Complement and its role in innate Y. Zhang, and M. H. Zou. 2012. Activation of AMP-activated protein kinase a2 and adaptive immune responses. Cell Res. 20: 34–50. by nicotine instigates formation of abdominal aortic aneurysms in mice in vivo. 17. Pagano, M. B., H. F. Zhou, T. L. Ennis, X. Wu, J. D. Lambris, J. P. Atkinson, Nat. Med. 18: 902–910. R. W. Thompson, D. E. Hourcade, and C. T. Pham. 2009. Complement- 37. Zhou, H. F., H. Yan, P. Bertram, Y. Hu, L. E. Springer, R. W. Thompson, dependent neutrophil recruitment is critical for the development of elastase- J. A. Curci, D. E. Hourcade, and C. T. Pham. 2013. Fibrinogen-specific antibody induced abdominal aortic aneurysm. Circulation 119: 1805–1813. induces abdominal aortic aneurysm in mice through complement lectin pathway 18. Zhou, H. F., H. Yan, C. M. Stover, T. M. Fernandez, S. Rodriguez de Cordoba, activation. Proc. Natl. Acad. Sci. USA 110: E4335–E4344. W. C. Song, X. Wu, R. W. Thompson, W. J. Schwaeble, J. P. Atkinson, et al. 38. Niinimaki, E., T. Paavonen, T. Valo, M. Tarkka, and A. Mennander. 2012. Lack 2012. Antibody directs properdin-dependent activation of the complement al- of C4d deposition may reveal susceptibility for ascending aortic dissection. ternative pathway in a mouse model of abdominal aortic aneurysm. Proc. Natl. Scand. Cardiovasc. J. 46: 177–182. Acad. Sci. USA 109: E415–E422. 39. Peng, Q., K. Li, S. H. Sacks, and W. Zhou. 2009. The role of anaphylatoxins C3a 19. Markiewski, M. M., and J. D. Lambris. 2007. The role of complement in in- and C5a in regulating innate and adaptive immune responses. Inflamm. Allergy flammatory diseases from behind the scenes into the spotlight. Am. J. Pathol. Drug Targets 8: 236–246. 171: 715–727. 40. Coulthard, L. G., and T. M. Woodruff. 2015. Is the complement activation 20. Bosmann, M., and P. A. Ward. 2012. Role of C3, C5 and anaphylatoxin receptors product C3a a proinflammatory molecule? Re-evaluating the evidence and the in acute lung injury and in sepsis. Adv. Exp. Med. Biol. 946: 147–159. myth. J. Immunol. 194: 3542–3548. 21. Jia, L. X., W. M. Zhang, H. J. Zhang, T. T. Li, Y. L. Wang, Y. W. Qin, H. Gu, and 41. Takabayashi, T., E. Vannier, B. D. Clark, N. H. Margolis, C. A. Dinarello, J. Du. 2015. Mechanical stretch-induced endoplasmic reticulum stress, apoptosis J. F. Burke, and J. A. Gelfand. 1996. A new biologic role for C3a and C3a and inflammation contribute to thoracic aortic aneurysm and dissection. J. desArg: regulation of TNF-alpha and IL-1 beta synthesis. J. Immunol. 156: Downloaded from Pathol. 236: 373–383. 3455–3460. 22. Humbles, A. A., B. Lu, C. A. Nilsson, C. Lilly, E. Israel, Y. Fujiwara, 42. Asgari, E., G. Le Friec, H. Yamamoto, E. Perucha, S. S. Sacks, J. Ko¨hl, N. P. Gerard, and C. Gerard. 2000. A role for the C3a anaphylatoxin receptor in H. T. Cook, and C. Kemper. 2013. C3a modulates IL-1b secretion in human the effector phase of asthma. Nature 406: 998–1001. monocytes by regulating ATP efflux and subsequent NLRP3 inflammasome 23. Lian, H., L. Yang, A. Cole, L. Sun, A. C. Chiang, S. W. Fowler, D. J. Shim, activation. Blood 122: 3473–3481. J. Rodriguez-Rivera, G. Taglialatela, J. L. Jankowsky, et al. 2015. NFkB- 43. Wu, M. C., F. H. Brennan, J. P. Lynch, S. Mantovani, S. Phipps, R. A. Wetsel, activated astroglial release of complement C3 compromises neuronal morphol- M. J. Ruitenberg, S. M. Taylor, and T. M. Woodruff. 2013. The receptor for

ogy and function associated with Alzheimer’s disease. Neuron 85: 101–115. complement component C3a mediates protection from intestinal ischemia- http://www.jimmunol.org/ 24. Botnar, R. M., A. J. Wiethoff, U. Ebersberger, S. Lacerda, U. Blume, A. Warley, reperfusion injuries by inhibiting neutrophil mobilization. Proc. Natl. Acad. C. H. Jansen, D. C. Onthank, R. R. Cesati, R. Razavi, et al. 2014. In vivo as- Sci. USA 110: 9439–9444. sessment of aortic aneurysm wall integrity using elastin-specific molecular 44. Daffern, P. J., P. H. Pfeifer, J. A. Ember, and T. E. Hugli. 1995. C3a is a che- magnetic resonance imaging. Circ Cardiovasc Imaging 7: 679–689. motaxin for human eosinophils but not for neutrophils. I. C3a stimulation of 25. Zhang, C., Y. Li, C. Wang, Y. Wu, W. Cui, T. Miwa, S. Sato, H. Li, W. C. Song, neutrophils is secondary to eosinophil activation. J. Exp. Med. 181: 2119–2127. and J. Du. 2014. Complement 5a receptor mediates angiotensin II-induced 45. Wang, L., J. Zhang, W. Fu, D. Guo, J. Jiang, and Y. Wang. 2012. Association of cardiac inflammation and remodeling. Arterioscler. Thromb. Vasc. Biol. 34: smooth muscle cell phenotypes with extracellular matrix disorders in thoracic 1240–1248. aortic dissection. J. Vasc. Surg. 56: 1698–1709, 1709.e1. 10.1016/j. 26. van der Maten, E., S. van Selm, J. D. Langereis, H. J. Bootsma, F. J. van jvs.2012.05.084 Opzeeland, R. de Groot, M. I. de Jonge, and M. van der Flier. 2016. Alternative 46. Li, W., Q. Li, Y. Jiao, L. Qin, R. Ali, J. Zhou, J. Ferruzzi, R. W. Kim, pathway inhibition by exogenous factor H fails to attenuate inflammation and A. Geirsson, H. C. Dietz, et al. 2014. Tgfbr2 disruption in postnatal smooth

vascular leakage in experimental pneumococcal sepsis in mice. PLoS One 11: muscle impairs aortic wall homeostasis. J. Clin. Invest. 124: 755–767. by guest on September 24, 2021 e0149307. 47. Wang, X., S. A. LeMaire, L. Chen, Y. H. Shen, Y. Gan, H. Bartsch, S. A. Carter, 27. Golovina, V. A., and M. P. Blaustein. 2006. Preparation of primary cultured B. Utama, H. Ou, J. S. Coselli, and X. L. Wang. 2006. Increased collagen de- mesenteric artery smooth muscle cells for fluorescent imaging and physiological position and elevated expression of connective tissue growth factor in human studies. Nat. Protoc. 1: 2681–2687. thoracic aortic dissection. Circulation 114(1 Suppl.): I200–I205. 28. Liu, Y., Y. Wang, H. Shi, L. Jia, J. Cheng, W. Cui, H. Li, P. Li, and J. Du. 2015. 48. Weis-Muller,€ B. T., O. Modlich, I. Drobinskaya, D. Unay, R. Huber, H. Bojar, CARD9 mediates necrotic smooth muscle cell-induced inflammation in mac- J. D. Schipke, P. Feindt, E. Gams, W. Muller,€ et al. 2006. Gene expression in acute rophages contributing to neointima formation of vein grafts. Cardiovasc. Res. Stanford type A dissection: a comparative microarray study. J. Transl. Med. 4: 29. 108: 148–158. 49. Song, Y., Y. Xie, F. Liu, C. Zhao, R. Yu, S. Ban, Q. Ye, J. Wen, H. Wan, X. Li, 29. Hinterseher, I., R. Erdman, L. A. Donoso, T. R. Vrabec, C. M. Schworer, et al. 2013. Expression of matrix metalloproteinase-12 in aortic dissection. BMC J. H. Lillvis, A. M. Boddy, K. Derr, A. Golden, W. D. Bowen, et al. 2011. Role of Cardiovasc. Disord. 13: 34. complement cascade in abdominal aortic aneurysms. Arterioscler. Thromb. Vasc. 50. Shaw, A., and Q. Xu. 2003. Biomechanical stress-induced signaling in smooth Biol. 31: 1653–1660. muscle cells: an update. Curr. Vasc. Pharmacol. 1: 41–58. 30. Martinez-Pinna, R., J. Madrigal-Matute, C. Tarin, E. Burillo, M. Esteban-Salan, 51. Zampetaki, A., Z. Zhang, Y. Hu, and Q. Xu. 2005. Biomechanical stress induces C. Pastor-Vargas, J. S. Lindholt, J. A. Lopez, E. Calvo, M. V. de Ceniga, et al. IL-6 expression in smooth muscle cells via Ras/Rac1-p38 MAPK-NF-kappaB 2013. Proteomic analysis of intraluminal thrombus highlights complement ac- signaling pathways. Am. J. Physiol. Heart Circ. Physiol. 288: H2946–H2954. tivation in human abdominal aortic aneurysms. Arterioscler. Thromb. Vasc. Biol. 52. Ignatius, A., P. Schoengraf, L. Kreja, A. Liedert, S. Recknagel, S. Kandert, 33: 2013–2020. R. E. Brenner, M. Schneider, J. D. Lambris, and M. Huber-Lang. 2011. Com- 31. Wu, G., T. Chen, A. Shahsafaei, W. Hu, R. T. Bronson, G. P. Shi, J. A. Halperin, plement C3a and C5a modulate osteoclast formation and inflammatory response H. Aktas, and X. Qin. 2010. Complement regulator CD59 protects against an- of osteoblasts in synergism with IL-1b. J. Cell. Biochem. 112: 2594–2605. giotensin II-induced abdominal aortic aneurysms in mice. Circulation 121: 53. Zhang, X., Y. Kimura, C. Fang, L. Zhou, G. Sfyroera, J. D. Lambris, 1338–1346. R. A. Wetsel, T. Miwa, and W. C. Song. 2007. Regulation of Toll-like receptor- 32. Ren, W., Y. Liu, X. Wang, L. Jia, C. Piao, F. Lan, and J. Du. 2016. b-Amino- mediated inflammatory response by complement in vivo. Blood 110: 228–236. propionitrile monofumarate induces thoracic aortic dissection in C57BL/6 mice. 54. Gutzmer, R., M. Lisewski, J. Zwirner, S. Mommert, C. Diesel, M. Wittmann, Sci. Rep. 6: 28149. A. Kapp, and T. Werfel. 2004. Human monocyte-derived dendritic cells are 33. Reiser, K., R. J. McCormick, and R. B. Rucker. 1992. Enzymatic and nonen- chemoattracted to C3a after up-regulation of the C3a receptor with interferons. zymatic cross-linking of collagen and elastin. FASEB J. 6: 2439–2449. Immunology 111: 435–443. Supplemental Table I Ascending

aorta Patient Age Gender Diagnosis Surgery diameter

(mm) Ascending aorta replacement + partial arch replacement + TAD1 (B) 61 Male 43 TAD noncoronary sinus plasty Ascending aorta replacement + total arch replacement by a TAD2 (B) 51 Male 46 TAD tetrafurcate graft and stented elephant trunk implantation Bentall + total arch replacement by a tetrafurcate graft and TAD3 (B) 43 Male 52 TAD stented elephant trunk implantation Bentall + total arch replacement by a tetrafurcate graft and TAD4 (B) 46 Male 40 TAD stented elephant trunk implantation Ascending aorta replacement + total arch replacement by a TAD5 (B) 60 Female 44 TAD tetrafurcate graft and stented elephant trunk implantation TAD6 (B) 38 Male 52 TAD Bentall Ascending aorta replacement + total arch replacement by a TAD7 (B) 45 Male 47 TAD tetrafurcate graft and stented elephant trunk implantation Bentall + total arch replacement by a tetrafurcate graft and TAD8 (B) 37 Male 56 TAD stented elephant trunk implantation Ascending aorta replacement + total arch replacement by a TAD9 (B) 67 Female 53 TAD tetrafurcate graft and stented elephant trunk implantation Ascending aorta replacement + total arch replacement by a TAD10 (B) 36 Male 64 TAD tetrafurcate graft and stented elephant trunk implantation Bentall + total arch replacement by a tetrafurcate graft and TAD11 (B) 43 Male 42 TAD stented elephant trunk implantation Ascending aorta replacement + total arch replacement by a TAD12 (B) 44 Male 45 TAD tetrafurcate graft and stented elephant trunk implantation Ctrl1 (B) 34 Male Unknown Healthy Null Ctrl2 (B) 46 Male Unknown Healthy Null Ctrl3 (B) 45 Male Unknown Healthy Null Ctrl4 (B) 52 Female Unknown Healthy Null Ctrl5 (B) 60 Female Unknown Healthy Null Ctrl6 (B) 38 Male Unknown Healthy Null Ctrl7 (B) 40 Male Unknown Healthy Null Ascending A replacement+ total arch replacement by a TAD1 (A) 59 Male 43 TAD tetrafurcate graft and stented elephant trunk implantation Ascending aorta replacement+ total arch replacement by a TAD2 (A) 44 Male 40 TAD tetrafurcate graft and stented elephant trunk implantation Ascending aorta replacement+ total arch replacement by a TAD3 (A) 51 Male 41 TAD tetrafurcate graft and stented elephant trunk implantation Ascending aorta replacement+ total arch replacement by a TAD4 (A) 41 Female 45 TAD tetrafurcate graft and stented elephant trunk implantation Ascending aorta replacement+ total arch replacement by a TAD5 (A) 50 Female 35 TAD tetrafurcate graft and stented elephant trunk implantation Ctrl1 (A) 55 Male 32 Recipient heart transplantation surgery Ctrl2 (A) 44 Male 27 Recipient heart transplantation surgery Ctrl3 (A) 22 Male Unknown Aorta DonorNon-cardiogenic death Ctrl4 (A) 73 Female Unknown Recipient heart transplantation surgery Ctrl5 (A) 47 Female Unknown Recipient heart transplantation surgery Supplemental Table I. Basic demographics of thoracic aortic dissection patients and controls. B: blood sample; A: aortic sample. Supplemental Table II

Gene Name Forward (5’-3’) Reverse (5’-3’)

mouse properdin tgcaaaggcctacttgggag-3’ tgaccattgtggagacctgc

mouse C1q ttcctcattttcccctcggc gacacagacggggatcgttt

mouse Mmp2 cgatgtcgcccctaaaacag gcatggtctcgatggtgttc

human C3aR caggactcgtggagacatcc aagacgccattgctaaacaaaaa

human MMP2 gagtgcatgaaccaaccagc aaacttgcagggctgtcctt

human MMP9 tctatggtcctcgccctgaa ttgtatccggcaaactggct

GAPDH acccagaagactgtggatgg cacattgggggtaggaacac

Supplemental Table II. Primers used in this study.

Supplemental Figure 1. Neutrophil infiltration in aortas. Immunofluorescence staining of neutrophils in aortas from WT control mice and WT and C3aR KO mice treated with

BAPN for 4 weeks with or without dissection. Alexa Fluor 488- labeled secondary antibody was used. Nuclei were stained with DAPI; n=6. Scale bar: 100 μm.

Supplemental Figure 2. Knockout of C3aR attenuated MMP9 expression in the aortas of BAPN-treated mice. The mRNA levels of MMP9 in aortas of 4-week of

BAPN-treated mice was assessed with qPCR and expressed as fold induction compared with the control after normalized to GAPDH. Data represent the mean±SEM; n=6.