Introduction

Cardiovascular diseases (CVDs) including atherosclerosis are a major cause of death and morbidity worldwide [1]. In China, 45.0% of deaths in rural area and 42.6% of deaths in urban area were caused by CVDs [2]. Although researchers have made great advances in diagnostics and therapy, CVDs still remain the leading cause of death globally. Thus, it is urgent to understand the potential molecular and cellular mechanisms involved in the pathological process of CVDs.

Innate immunity is the first barrier to defend against infection which is also associated with a number of sterile inflammatory conditions such as CVDs [3]. Toll-like receptors (TLRs) and NOD-like receptors (NLRs) are two main sensor families in the innate immune system to detect pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) [4, 5]. NOD2 localized in the cytoplasm was one of first receptors characterized in the NLRs family [6]. According to research, NOD2 has recently been reported to be a central regulator of immune and inflammatory responses in CVDs, such as atherosclerosis and cerebral ischemia/reperfusion injury [5, 7]. Muramyl dipeptide (MDP), a peptidoglycan constituent of both Gram-positive and Gram-negative bacteria, is a defined cognate ligand of NOD2. Peptidoglycan has been detected in atherosclerotic plaques and likely derives from gut microbiota [8]. Furthermore, NOD2 regulates microbiota communities in the gut [9]. Thus, it is important and necessary for us to study MDP-triggered NOD2 innate immune responses in CVDs.

CVDs develop due to a complex interplay of inflammation and coagulation, in which endothelial cells (ECs) are considered to be a central regulator [10, 11]. Endothelial injury leads to the release of coagulation and inflammation mediators, accelerating the adhesiveness of leukocytes [11, 12]. Vascular cell adhesion molecule‑1 (VCAM-1) and intercellular cell adhesion molecule‑1 (ICAM-1), which are both members of the immunoglobulin gene superfamily, are overexpressed in the pathogenesis of endothelial dysfunction [13]. VCAM‑1 is capable of mediating molecular adhesion to ECs after induction of inflammatory cytokines [14]. Numerous studies have reported VCAM‑1 and ICAM‑1 serve as markers of early atherosclerosis and play a significant role in development of atherosclerosis [15,16,17,18]. Furthermore, it has been demonstrated that MDP induces mRNA expression of ICAM‑1 in ECs [8], but whether VCAM‑1 expression could be affected by MDP is not fully understood. ET‑1 synthesized and secreted by ECs has been recognized as the most potent endogenous vasoconstrictor [19]. It accelerates the pathophysiologic progress of CVDs via vasoconstriction. Studies have indicated that atherosclerosis is aggravated by the ET-1-triggered NF-κB signaling pathway [20, 21]. Moreover, there have been reports that lipopolysaccharides (LPS), the principal component of Gram-negative bacteria, dose-dependently increase the release and expression levels of ET‑l in HUVECs [22]. However, it remains unanswered whether MDP leads to a change in VCAM‑1 and ET‑1 production in ECs. Since NOD2 is the defined cognate ligand receptor of MDP, triggered NF-κB-dependent transcriptional activity in ECs after MDP stimulation [23], suggested NOD2 played a potential important role in the ECs. Furthermore, we previously proved MDP induced oxidative stress through the NOD2/COX-2/NOX4 signaling pathway in the HUVECs [6]. However, there was no direct evidence whether NOD2 is involved in the production of VCAM‑1 and ET‑1. Thus, in this study, our goal was to investigate the role and mechanism of NOD2 in MDP-triggered VCAM‑1 and ET‑1 gene expression in primary human umbilical vein endothelial cells (HUVECs).

Materials and methods

Cell culture and stimulation

The cell line of HUVECs was purchased from ScienCell Research Laboratories (Carlsbad, CA, USA) which provides a variety of normal human and animal cells for the research community. Briefly, the HUVECs were cultured with endothelial cell medium (ECM, ScienCell Research Labs, Carlsbad, CA, USA) supplemented with 5% fetal bovine serum, 1% penicillin–streptomycin, and 1% endothelial cell growth supplement (ECGS) at 37 °C with 5% CO2.

Reagents

MDP (53678-77‑6, InvivoGen, San Diego, CA, USA) was used at a final concentration of 10 μg/mL. Pyrrolidine dithiocarbamate (PDTC) was obtained from Beyotime Biotechnology (Shanghai, China).

Small interfering RNA and cell transfection

Scrambled control small interfering RNA (siRNA) and NOD2 siRNA were purchased from Sangon Biotech Co. Ltd. (Shanghai, China). The sequences of NOD2-targeted siRNA and non-targeted siRNA are listed in Table 1. The HUVECs were seeded in a 6-well plate at a density of 4 × 105 cells per well and cultured overnight to 70% confluence. Scrambled control siRNA and NOD2 siRNA were transfected into the HUVECs at a final dose of 50 nM diluted in the ECM medium without fetal bovine serum by Lipofectamine 2000 reagent (Invitrogen, Waltham, MA, USA), following the manufacturer’s instructions.

Table 1 Sequence of small interfering RNA (siRNA) sets for cell transfection
Full size table

RNA extraction and real-time reverse transcription (qRT) PCR

Total RNA was extracted from HUVECs using trizol reagent (Invitrogen, Waltham, MA, USA). RT-PCR was implemented by a prime script RT reagent kit (Takara, Dalian, China) according to the manufacturer’s instruction. The mRNA levels of VCAM‑1 and ET‑1 were measured by qRT-PCR using SYBR premix ex taq II (Takara, Dalian, China) on a Bio-Rad iCycler system (BioRad, Hercules, CA, USA). The 2−∆∆CT method was adopted to determine the relative expression of target genes. The specific primers for qRT-PCR are listed in Table 2. GAPDH was used as the housekeeping gene.

Table 2 List of primer sets for SYBR-Green-based real-time reverse transcription (qRT) PCR
Full size table

Statistics

All of the experiments were performed in triplicate. Data were expressed as mean ± SEM and analyzed with the student’s t‑test using GraphPad Prism 5.01 software. A value of P < 0.05 was considered statistically significant.

Results

MDP up-regulated mRNA expression levels of VCAM-1 and ET-1

In order to examine the effect of MDP in ECs, we treated HUVECs with MDP for 12 h at a concentration of 10 μg/mL. From the experimental data shown in Fig. 1, both VCAM‑1 and ET‑1 mRNA expression were significantly up-regulated by MDP, especially VCAM‑1. The mRNA expression level of VCAM‑1 in the MDP-treated group was more than twice as high as the control group. These data suggest that NOD2, the MDP specific receptor, might play an important role in VCAM‑1 and ET‑1 gene expression in HUVECs.

Fig. 1
figure 1

Gene expression levels of VCAM‑1 (a) and ET‑1 (b) were promoted by MDP in HUVECs. HUVECs were cultured with MDP for 12 h at an indicated dose of 10 μg/mL. The mRNA expression of VCAM‑1 and ET‑1 in HUVECs was detected via qPCR. *P < 0.05, **P < 0.01 and ***P < 0.001, vs. control. VCAM-1 vascular cell adhesion molecule‑1, ET‑1 endothelin‑1, HUVEC human umbilical vein endothelial cells, MDP muramyl dipeptide, qPCR quantitative polymerase chain reaction

Full size image

NOD2 gene silence inhibited VCAM-1 and ET-1 gene expression in MDP-treated HUVECs

To assess the exact function of NOD2 in the MDP-induced increase of VCAM‑1 and ET‑1 mRNA levels, NOD2-siRNA was transfected into HUVECs with Lipofectamine 2000 regent according to the manufacturer’s instructions. Our results showed that there was no statistical difference between the two groups regarding NOD2 siRNA or NOD2 siRNA with MDP stimulation, based on the gene expression of VCAM‑1 and ET‑1 (Fig. 2). After NOD2 knockdown, VCAM‑1 and ET‑1 gene expression were both inhibited in MDP-treated HUVECs. Thus, it confirmed that NOD2 plays a critical role in VCAM‑1 and ET‑1 gene up-regulation.

Fig. 2
figure 2

NOD2 inhibition suppressed mRNA expression of VCAM‑1 (a) and ET‑1 (b) in MDP-treated HUVECs. Scrambled siRNA and NOD2 siRNA respectively were transferred into HUVECs. VCAM‑1 and ET‑1 mRNA expression were determined by qPCR. *P < 0.05, **P < 0.01 and ***P < 0.001, vs. control. NOD2 nucleotide-binding oligomerization domain-containing protein 2, VCAM-1 vascular cell adhesion molecule‑1, ET‑1 endothelin‑1, siRNA small interfering RNA, HUVEC human umbilical vein endothelial cells, MDP muramyl dipeptide, qPCR quantitative polymerase chain reaction

Full size image

VCAM-1 and ET-1 mRNA expression levels down-regulated by PDTC in MDP-treated HUVECs

NF-κB plays a regulatory role in immune response and participates in inflammatory cytokine production [24, 25]. It was reported that LPS enhances expression of VCAM‑1 via the NF-κB signaling pathway in HUVECs [8]. Furthermore, ET‑1 expression was up-regulated by NF-κB in LPS-treated ECs [22]. Therefore, we speculated that MDP induces VCAM‑1 and ET‑1 gene expression via NF-κB. Next, the HUVECs were pre-incubated with PDTC, which is a specific inhibitor of NF-κB, for 2 h before MDP stimulation. Results indicated that there was a significant difference between the two conditions (Fig. 3). Compared to the MDP group, mRNA expression of VCAM‑1 and ET‑1 were dramatically suppressed in MDP-cultured HUVECs treated with PDTC. Taken together, it was noted that NF-κB is a critical mediator in NOD2-triggered VCAM‑1 and ET‑1 gene overexpression in HUVECs (Fig. 4).

Fig. 3
figure 3

PDTC suppressed the mRNA expression of VCAM‑1 (a) and ET‑1 (b) in MDP-treated HUVECs. HUVECs were pretreated with PDTC at a concentration of 100 μM for 2 h, and then stimulated with MDP for 12 h. qPCR was performed to detect the mRNA expression of VCAM‑1 and ET‑1. *P < 0.05, **P < 0.01 and ***P < 0.001, vs. control. PDTC pyrrolidine dithiocarbamate, VCAM-1 vascular cell adhesion molecule‑1, ET‑1 endothelin‑1, HUVEC human umbilical vein endothelial cells, MDP muramyl dipeptide, qPCR quantitative polymerase chain reaction

Full size image
Fig. 4
figure 4

Schematic representation showing the NOD2 signaling pathway in HUVECs. MDP stimulation promoted gene expression of VCAM‑1 and ET‑1 in endothelial cells. VCAM-1 vascular cell adhesion molecule‑1, ET‑1 endothelin‑1, HUVEC human umbilical vein endothelial cells, MDP muramyl dipeptide, qPCR quantitative polymerase chain reaction, NF-κB nuclear factor κB

Full size image

Discussion

Although research into NOD2-mediated innate immune responses in the pathogenesis of inflammatory diseases has recently made great progress, our research provides new insights on NOD2 in CVDs. Vascular endothelium cells are now acknowledged to be a major regulator of vascular homeostasis [26, 27]. ECs dysfunction is a significant marker of CVDs [28]. In the present study, we revealed that MDP, which is regarded as an agonist of NOD2, induced mRNA expression of VCAM‑1 and ET‑1 through NF-κB in the HUVECs.

The innate immune response has been reported to be definitely involved in the chronic inflammation of CVDs [5, 6, 29]. NOD2 which belongs to the NOD-like receptor family is an important mediator in the innate immune system. Growing evidence has confirmed that NOD2 activation triggers the caspase recruitment domain (CARD) and then induces the NF-κB signaling pathway in inflammatory diseases [3, 5, 30]. Compared with the low expression of NOD2 in healthy arteries, its abundance in lesions signifies that the activity of human atherosclerosis is associated with enhanced NOD2-mediated innate immunity [5]. A previous study noted that NOD2 is minimally expressed and readily induced by proinflammatory cytokines in ECs [23]. It indicated that NOD2 is a potential target for the modulation of vascular inflammation. However, there is still doubt of how NOD2 is activated in ECs. An emerging view has been that the gut microbiota is likely the source of peptidoglycans in atherosclerosis. Bacterial phylotypes in atherosclerotic plaques are common to gut microbiota [8]. In several reports, peptidoglycans in vulnerable atherosclerotic plaques were found more often than in stable plaques [5], while MDP, known as a component of most bacterial peptidoglycan, is the only specific ligand for NOD2. Thus, we suggested that MDP might trigger the NOD2 signal pathway in ECs. Moreover, NOD2 plays an important role in the balance of gut microbiota communities [9]. Based on these studies, our goal was to mainly focus on the mechanism of NOD2-induced expression of ET-1 and VCAM-1 rather than other pro-oxidant cytokines.

VCAM‑1, which functions as the mediator of immune cells migration, is a major member in the immunoglobulin superfamily of adhesion molecules [31]. ET‑1 produced by vascular ECs is an amino acid vasoconstrictor peptide and is mainly expressed in the cardiovascular system [32]. It has been demonstrated that ET‑1 overexpression leads to vascular oxidative stress and inflammation, which significantly contributes to development of atherosclerosis [32,33,34]. Our results showed that mRNA expression levels of VCAM‑1 and ET‑1 are up-regulated in HUVECs cultured with MDP for 12 h, which suggests that NOD2 in vascular endothelium might be a potential regulator in adhesion molecules and ET-1expression. Since VCAM‑1 and ICAM‑1 are both endothelial adhesion molecules in the Ig gene superfamily, this is consistent with the report that low expression of NOD2 enabled ECs to express ICAM‑1 in response to MDP [8]. In support of our suggestion, we also investigated the role of NOD2 in MDP-induced VCAM‑1 and ET‑1 expression. The data illustrate that NOD2 indeed mediates the gene levels of VCAM‑1 and ET‑1 in HUVECs stimulated with MDP. As cells lacking NOD2 showed little VCAM‑1 and ET‑1 gene expression, we speculated it might be supplemented by other signaling pathways. Furthermore, because there was no significant difference between the two NOD2-knockdown groups with MDP stimulation or without, it suggested that NOD2 is one of the most critical inducers.

Nevertheless, it still remains unknown how NOD2 mediates VCAM‑1 and ET‑1 overexpression. Coincidentally, we noticed that NOD2-mediates IL‑6, IL‑8, and VCAM‑1 expression in the Porphyromonas gingivalis-induced NF-κB signaling pathway [35]. Previous studies indicated that NOD2 induces NF-κB-dependent transcriptional activity in HUVECs [23]. NF-κB is an important modulator of ET‑1 production in the ECs [36]. In addition, NF-κB participates in the inflammatory signaling pathways in CVDs, leading to the increase of VCAM‑1 expression and monocyte adhesion [37, 38]. Therefore, we investigated whether NF-κB was involved in MDP-induced VCAM‑1 and ET‑1 gene overexpression in HUVECs. As expected, our results from this investigation showed that mRNA levels of VCAM‑1 and ET‑1 are suppressed by PDTC which serves as a NF-kB inhibitor in HUVECs stimulated with MDP. These investigations are in accordance with the report that the expression levels of VCAM‑1 and ET‑1 are up-regulated after LPS treatment [39]. Thus, we confirmed that NF-κB plays a pivotal role in the NOD2-mediated VCAM‑1 and ET‑1 gene expression in HUVECs.

Overall, this study concentrated on NOD2-mediated VCAM‑1 and ET‑1 gene expression in ECs, thus, contributing to our understanding of the innate immune mechanism of ECs. However, since protein levels reveal the regulatory mechanism more accurately and rigorously, it would be more informative and convincing if the protein expression data of VCAM‑1 and ET‑1 were provided. As significant exploration of NOD2 in the pathological process of ECs is needed, we will be conducting these protein experiments in the future.

Conclusion

We confirmed that MDP stimulation triggered mRNA overexpression of VCAM‑1 and ET‑1 in HUVECs. Although NOD2 expression in the ECs was low, it was a significant modulator in VCAM‑1 and ET‑1 mRNA expression. Furthermore, NOD2 mediated VCAM‑1 and ET‑1 gene expression via NF-κB in HUVECs treated with MDP.