Serum Amyloid A Promotes E-Selectin Expression via Toll-Like Receptor 2 in Human Aortic Endothelial Cells

Periodontitis is a chronic inflammatory disease that affects the periodontium. Recent studies suggest an association between periodontal and cardiovascular diseases. However, the detailed molecular mechanism is unknown. A previous study has demonstrated that experimental periodontitis induces serum amyloid A (SAA) in the liver and peripheral blood of ApoE-deficient mice as an atherosclerosis model. SAA is an acute-phase protein that affects systemic inflammation. The aim of this study is to investigate the atherosclerosis-onset mechanism using human aortic endothelial cells (HAECs) stimulated by SAA in vitro. Atherosclerosis PCR array and qPCR analyses showed upregulation of adhesion molecules such as intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin in HAECs upon SAA stimulation. In addition, the results demonstrated that Toll-like receptor, TLR2, could serve as an important receptor of SAA in HAECs. Furthermore, small interfering RNA (siRNA) against TLR2 inhibited the upregulation of adhesion molecules in HAECs stimulated by SAA. Our results suggest that SAA stimulates the expression of adhesion molecules via TLR2. SAA could be an important molecule for atherosclerosis induced by periodontal disease.

The prevalence and incidence of coronary heart disease are significantly increased in patients with periodontitis, indicating that periodontal disease independently predicts coronary heart disease [10]. Recent systematic reviews and meta-analysis of observational studies support an association between periodontal and atherosclerotic vascular diseases, which is independent of known confounders, but a causal relationship is not yet established [11,12].
Serum amyloid A (SAA), an acute-phase protein, is markedly upregulated in response to infection and during chronic inflammation [13][14][15]. SAA stimulates vascular cells to express cytokines, chemokines, adhesion molecules, and matrix metalloproteinases [16], which are linked to the development of atherosclerosis. In addition, high levels of SAA in peripheral blood are significantly associated with periodontitis, and SAA levels are decreased in patients with periodontitis after periodontal therapy [17,18].
Recently, evidence of the possible link between periodontitis and atherosclerosis has increased, and the possible association and causality are being investigated [19,20]. Hujoel 2 Mediators of Inflammation  CCC TCG ATT CAA TTG CCT TA-3  Antisense  5 -TGT GAC CAA TGA CCT CAT GC-3   ABCA1  Sense  5 -TTT GCT GTA TGG GTG GTC AA-3  Antisense  5 -AAC AGC TCC AGC ACA AAG GT-3   ABCA7  Sense  5 -ATG TGG TGC TCA CCT GCA TA-3  Antisense  5 -AAG CAG AAG TGG GGG AAG AT-3   SCARB1  Sense  5 -CTC CCA TCC TCA CTT CCT CA-3  Antisense  5 -GCT CAG CTG CAG TTT CAC TCC TTG AAT GCC TCA AT-3   AGER  Sense  5 -CTG AGG CAG GCG AGA GTA GT-3  Antisense  5 -TTG GCA AGG TGG GGT TAT AC-3   GAPDH  Sense  5 -GTC AGT GGT GGA CCT GAC CT-3  Antisense  5 -TCG CTG TTG AAG TCA GAG GA-3 et al. reported an important general trend towards periodontal treatment-induced inhibition of systemic inflammation and improvement in noninvasive markers of atherosclerosis and endothelial function [19]. However, the detailed molecular mechanism for periodontitis-induced atherosclerosis is unknown. The aim of this study is to investigate the atherosclerosisonset mechanism using human aortic endothelial cells (HAECs) stimulated by SAA in vitro. Here, we demonstrate that stimulation of HAECs with SAA results in the induction of adhesion molecules, which may be caused via Toll-like receptor, TLR2.

Cell
Culture and SAA Treatment. HAECs were purchased from Lonza (CC-2535; Tokyo, Japan) and used in all experiments. The HAECs were cultured in Endothelial Cell Growth Medium 2 (EBM-2) medium supplied with the EBM-2 bullet kit (Lonza, Tokyu, Japan) at 37 ∘ C with 5% CO 2 . Subconfluent passage 6 cells were used in all experiments. HAECs were plated at 1.5 × 10 5 cells/well in 6-well plates and cultured to subconfluence. The cells were then treated with 1.5 g/mL recombinant human SAA (PeproTech, Rocky Hill, NJ) for 0, 1, 3 and 6 h.

PCR Array Analysis.
Total RNA was extracted from HAECs using NucleoSpin RNA II (Macherey-Nagel, Diiren, Germany). First-strand cDNA synthesis was performed using RT 2 First Strand Kit (Qiagen, Tokyo, Japan) following the manufacturer's instructions. The Human Atherosclerosis RT 2 Profiler6 PCR Array (PAHS-038Z) (Qiagen, Tokyo, Japan) was applied to an ABI 7000 Real-Time PCR System (Applied Biosystems, Foster City, CA). The RT 2 Profiler6 PCR Array for Human Atherosclerosis contains 84 genes for responses to stress, apoptosis, blood coagulation and circulation, adhesion molecules, extracellular molecules, lipid transport and metabolism, and cell growth and proliferation. In addition, the array contains five wells for various housekeeping genes, a genomic DNA contamination control, three replicate reverse transcription controls, and three replicate positive PCR controls. Data analyses were performed using web-based analysis software (http://pcrdataanalysis.sabiosciences.com/ pcr/arrayanalysis.php).

RNA Interference.
HAECs were transfected with TLR2 Silencer Select Pre-Designed Small Interfering RNA (siRNA, ASO0ZS41; Life Technology, Tokyo, Japan) using Lipofectamine 2000 transfection reagent (Life Technology, Tokyo, Japan) according to the manufacturer's instructions. HAECs (1.5 × 10 5 cells/well) were seeded in 6-well plates at 24 h before transfection. Then, the cells were transfected with siRNA against TLR2 at a final concentration of 10 nM. Cells were incubated with siRNA in EBM-2 medium for 24 h. Then, the medium was replaced with fresh medium, and the cells were incubated for another 48 h. HAECs were then stimulated with SAA before harvesting. TLR2 knockdown was verified by qPCR. Control siRNA was purchased from Life Technology.

Statistical Analysis.
Statistical analyses were performed using SPSS software v. 15.0 J for Windows (SPSS Inc., Chicago, IL). Data are expressed as the mean ± standard deviation. Student's t-test was used for comparisons. Significance was accepted at < 0.05.

TLR2 Is Upregulated by SAA among Receptor Molecules in HAECs.
To identify genes related to the leukocyte adhesion cascade, we screened SAA receptors that were highly expressed in HAECs during SAA stimulation ( Figure 2). SAA receptors, such as SELS (glucose homeostasis and ER stress), ABCA1, ABCA7, SCARB1 (cholesterol efflux), CD36, TLR2, TLR4, CST3 (inflammatory signaling), FPR2 (chemotaxis and immune cell activation), and AGER (amyloidosis), have been reported previously [21]. Among the candidate receptors, TLR2 mRNA expression was significantly induced by SAA in HAECs, indicating that TLR2 could serve as an important receptor for SAA. Thus, SAA may stimulate the expression of adhesion molecules via TLR2.

SAA Induces TLR2 and Its Related Genes following the Leukocyte Adhesion Cascade.
To investigate the leukocyte adhesion cascade induced by SAA, mRNA expression was examined at 0, 1, 3, and 6 h after SAA stimulation in HAECs ( Figure 3). TLR2 mRNA expression was upregulated in a time-dependent manner. Furthermore, the mRNA expression of NFKB1, TNF-, and SELE was significantly upregulated at 3 h after SAA stimulation. The mRNA expression of these genes was significantly higher in SAA-stimulated HAECs compared with unstimulated HAECs. However, only the mRNA expression of MYD88 was lower in SAAstimulated HAECs compared with unstimulated HAECs.
These results indicate that SAA affects downstream of TLR signaling and induces leukocyte adhesion cascade-related genes.

Knockdown of TLR2 Affects the Expression of SELE through the Leukocyte Adhesion Cascade.
To determine the contribution of TLR2 as a SAA receptor for SELE expression, TLR2 was knocked down by transfection of siRNA. We confirmed specific knockdown of TLR2 by qPCR and western blotting (Figures 4(a) and 4(b)). No difference in cell morphologies was found in HAECs transfected with TLR2 siRNA (Figure 4(c)). Silencing of TLR2 using siRNA resulted in a significant reduction in the expression of SELE mRNA compared with cells transfected with control siRNA under SAA stimulation ( Figure 5). In addition, mRNA expression of NFKB1 and TNF-under SAA stimulation was abolished by silencing TLR2.

Discussion
In this study, we found that SAA induced adhesion molecules, especially SELE, and TLR2 may be a critical receptor for this process in HAECs. To our knowledge, this is the first report indicating that TLR2 plays a role in HAECs as a SAA receptor 6  Relative expression/18S Relative expression/18S * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * to induce adhesion molecules, which might be involved in the onset of atherosclerosis. The mRNA expression of adhesion molecules, including ICAM1, VCAM1, and SELE, was dramatically upregulated (>5-fold) by SAA stimulation in HAECs (Figure 1). E-selectin is important for the initial rolling interactions of neutrophils, monocytes, natural killer cells, and a subset of memory T cells in the inflamed endothelium [22,23]. Leukocyte arrest during rolling is mediated by binding of leukocyte integrins to immunoglobulin superfamily members, such as VCAM-1 and ICAM-1, which are expressed by endothelial cells [24].

Mediators of Inflammation
The initial step of atherosclerosis includes adhesion of peripheral blood leukocytes to activate the endothelial monolayer, directed migration of the bound leukocytes into the intima, and maturation of monocytes into macrophages and their uptake of lipids, yielding foam cells [25]. Multiple members of selectin, integrin, and immunoglobulin gene families in the process of initial attachment (rolling), stable adhesion (arrest), spreading, and ultimately diapedesis are sequentially involved in the onset of atherosclerosis [25]. Based on the PCR array results, SAA induces adhesion molecules and thus may trigger arteriosclerosis onset.
Several SAA receptors have been reported previously [21]. Among them, the expression of TLR2 was dramatically increased (about 30-fold) during SAA stimulation (Figure 2). We confirmed upregulation of TLR2 at both mRNA and protein levels by SAA stimulation, and SAA increased the expression of SELE in conjunction with the leukocyte adhesion cascade in HAECs. Endothelial cells normally express TLR2 at a very low level [26]. Thus, our results suggest that TLR2 may be an important receptor for SAA to induce adhesion molecules in HAECs.
Based on qPCR analysis of downstream molecules in TLR2 signaling (Figure 3), SAA stimulation had a remarkable influence on their expression. In particular, expression of the leukocyte cascade (i.e., NFKB1 and TNF-) was upregulated following the induction of TLR2 by SAA stimulation. In addition, analysis of HAECs transfected with siRNA against TLR2 showed downregulation of the expression of the leukocyte cascade (NFKB1 and TNF-) and SELE in a time-dependent manner ( Figure 5). MyD88 has a critical role in signaling via TLR2. After stimulation, TLR recruits IL-1R-associated kinase via adaptor MyD88 and induces activation of mitogen-activated protein kinases and nuclear factor-B (NF-B) [27]. NF-B activation induces proinflammatory cytokines, including TNF-and chemokines [28]. Furthermore, in human endothelial cells, adhesion molecules such as E-selectin are induced by NF-B and TNF- [29]. Nevertheless, in the current study, mRNA expression of MYD88 was decreased in SAA-stimulated HAECs, and the mRNA expression of NFKB1, TNF-, and SELE was upregulated. We speculate that SAA may induce NF-B, TNF-, and E-selectin, in part, through a MyD88independent pathway via TLR2. Indeed, Nilsen et al. have recently reported a novel function of TRAM/TRIF in TLR2mediated signal transduction [30]. However, further studies are needed to investigate the detailed mechanism of MyD88dependent or MyD88-independent signaling pathways in HAECs treated with SAA.

Conclusions
Our results suggest that TLR2 may have a critical role to induce adhesion molecules following TLR2 signaling and the leukocyte cascade in SAA-stimulated HAECs ( Figure 6). And SAA might be a predictive risk marker for atherosclerosis onset in patients with periodontitis.