This article is mentioned in:

Introduction

Atherosclerosis is the major cause of cardiovascular disease (CVD), which is the leading cause of death in developed countries (1). During the last decade, increasing amount of evidence supports chronic systemic vascular inflammation as an essential requirement for the progression of atherosclerosis and that the innate and adaptive immune cells play an integral role (2). Activation of vascular endothelial and smooth muscle cells by several endogenous or exogenous moieties causes up regulation of inflammatory cytokines like interleukin (IL)-1β and IL-18, augment proliferation, accumulation of oxidized low density lipoprotein (oxLDL) and extracellular matrix (ECM) components by the activation of intracellular pattern-recognition receptors (PRRs) (3,4).

IL-1β, IL-1α and IL-1 receptor antagonist (Ra) are members of the IL-1 family protein encoded by genes located on chromosome 2 (5). IL-1Ra is known to be a natural competitor for similar binding sites with IL-1β, which is an important mediator of cellular processes such as differentiation, proliferation and apoptosis (5). IL-1β has been implicated for the recruitment of inflammatory cells to infection sites, while IL-18 is important for IFN-γ production and enhancement of natural killer cell activity (5,6). IL-1β and IL-18 are produced from their 31 and 24 kDa inactive forms to their 17 and 18 kDa bio-active forms, respectively, through a proteolytic cleavage by caspase-1 and this depends on a complex protein platform called the inflammasome (7–9).

Previous studies have identified the NLRP1, NLRP3, IPAF/NLRC4 and AIM2 inflammasomes to activate caspase-1 (10). These complex proteins consist of three major components: a receptor [nucleotide oligomerisation domain (NOD)-like receptor (NLR)], which acts as an intracellular sensor and induces complex formation; in most cases an adaptor protein [apoptosis speck-like protein containing (ASC) a caspase recruitment domain (CARD) and caspase-1] (10). It has been hypothesized that upon stimulation of the inflammasome receptors by exogenous and endogenous stimuli like pathogens, cellular stress, low K+ concentrations, crystals or uric acid, the receptors undergo conformational changes forming inflammasome complexes by either recruiting an ASC via their pyrin-pyrin interaction before binding to caspase-1, or binding directly to pro-caspase-1 through CARD-CARD interaction (9,11–13). This leads to activation of caspase-1, which in turn converts inactive IL-1β to its active form (14). Unlike caspase-1 and NLRs that are expressed in resting cells, IL-1β mRNA is believed to emanate from the activation of the NF-κB pathway; or microbial induction via Toll-like receptor, C-type lectin receptor or RIG-1-like receptor’s ligand binding; or through MyD88 activation via IL-1 receptor binding (10).

CARD8 is a member of the CARD family and has been implicated to be a co-regulator in both inflammatory and apoptotic signaling pathways, although their exact role in these processes remain elusive and thus needs to be further explored (15). The CARD8 gene is located on chromosome 19q13 (16) and polymorphisms in this gene have been associated with several auto-inflammatory diseases, such as inflammatory bowel disease (IBD) and rheumatoid arthritis (RA) (17,18). Five isoforms of CARD8 have been identified but a contradictory functional role has been reported between the 48- and 54-kDa isoforms (19). The 48-kDa isoform has been implicated to interact directly with caspase-1 to induce apoptosis, while the 54-kDa isoform suppresses caspase mediated apoptosis (19,20). The isoforms have been reported to show a different expression pattern in different tissues and tumor cell lines (19).

Concurrently, studies have been carried out to explore the functional role of NLRP3 and CARD8 mainly in inflammatory (monocytes and macrophages) and cancer cells, but less is known about their functional role in vascular cells. Due to the expression of IL-1 family proteins in vascular cells and in atherosclerotic plaque (21–23) together with detrimental effects associated with excess IL-1β, it is therefore crucial to understand how vascular cells sense infection and metabolic stress and initiate vascular cell inflammation. The aim of this study is to examine the actual role of NLRP3 inflammasome and CARD8 protein in IL-1β expression and release from aortic smooth muscle cells (AOSMCs).

Materials and methods

Materials

Human AOSMC, OPTI-MEM reduce serum medium, Lipofectamine 2000, AOSMC Basal medium (M-231–500) and specific primers for Tucan 47F, Tucan 48F, Tucan 54F and Tucan Ex-9-R (19); and TaqDNA polymerase (18038/042) were purchased from Invitrogen (Stockholm, Sweden), Universal negative control and specific siRNA’s for NLRP3 and CARD8 were obtained from Sigma-Aldrich (Stockholm, Sweden), E.Z.N.A. total-RNA kit I from Omega Biotech (Doraville, GA, USA), 2X TaqMan Universal PCR Master Mix and probes for gene expression analysis were from Applied Biosystems (Foster City, CA), tumor necrosis factor-α (TNF-α) from PeproTech (Stockholm, Sweden), IL-1β and IL-1Ra kit for ELISA was obtained from DuoSet® Development System (R&D Systems, UK). Primary antibodies against NLRP3 (MAB3724) and CARD8 (PAB0218) were obtained from Abnova Corp. (Taipei City, Taiwan). 5X Green Go reaction buffer (M719A) was purchased from Promega Biotech AB (Stockholm, Sweden).

Cell culture

Human AOSMC cells were grown in smooth muscle cell growth medium with medium change every second day. For experiments, passages 5-10 cells were used and 2.5x105 cells/well were seeded in 6-well plates and treated with or without TNF-α for 1, 6, 24 or 48 h. Cell culture supernatants were used for measuring IL-1β and IL-1Ra expression.

Knockdown of NLRP3 and CARD8

Knockdown of NLRP3 and CARD8 were performed on 2x105 cells/well seeded in 6-well plates. The transfection mixture was prepared by using 4 μl of Lipofectamine 2000 and 10 or 20 pmol of NLRP3 (Hs0200313821) or CARD8 (Hs0034180) siRNA, respectively, to 500 μl of Opti-MEM. The tubes were incubated for 20 min. The mixture was then added to cells and incubated for 8 h followed by addition of antibiotic free growth medium and incubated for 24 h. Media was then replaced and cells stimulated with 50 ng/ml TNF-α and incubated for further 24 h.

Quantitative real-time-polymerase chain reaction (qRT-PCR)

In accordance with the manufacturer’s instruction, total-RNA was extracted using the E.Z.N.A. total-RNA kit. Using random polyhexamers, RNaseOUT and Superscript II, 0.75 mg of RNA was reverse-transcribed into cDNA. For gene expression analysis, IL-1β (Hs001740097), NLRP3 (Hs00366465), IL-1RN (Hs00277299), CARD8 (Hs01088228) and cyclophilin A (Hs04194521) were used, cDNA was amplified in the 7900HT Fast Real-Time PCR system (Applied Biosystems) on fast optical 96-well plates according to the manufacturer’s instructions. Each sample was analyzed in duplicate and values obtained were normalized with cyclophilin A.

Enzyme-linked immunosorbent assay (ELISA)

We employed the ELISA technique for the detection of IL-1β and IL-1Ra secreted in cell culture supernatant using the DuoSet® Development System kits. In accordance with the manufacturer’s instruction with slight modification, 0.05% Tween-PBS was used as reagent diluent in place of 1% BSA. Samples were diluted in equal volumes with 0.05% Tween-PBS for IL-1β quantification, but not for IL-1Ra, and each sample was analyzed in duplicate.

Reverse transcription-polymerase chain reaction (RT-PCR)

cDNA obtained from reverse transcriptase as mentioned above was amplified in a 50 μl PCR mixture. Tucan-54F/Ex9-R, Tucan-48F/Ex9-R and Tucan-47F/Ex9-R were used as forward and reverse primer pairs as previously described (19). The PCR mixture was amplified as follows: 10 min at 94˚C followed by 40 cycles at 45 sec at 94˚C, 30 sec at 58˚C, and 90 sec at 72˚C. The cycling procedure was followed by 10 min extension at 72˚C. The PCR mixture was viewed under UV light post electrophoresis on an ethidium bromide stained 1% agarose gel.

Western blot analysis

Protein lysates from transfected ASOMC were subjected to electrophoresis on 8 or 12% SDS-polyacrylamide gels and further transfered to PVDF membrane. Membranes were blocked with 5% non-fat milk in 0.01% Tween-PBS with shaking for 4 h and incubated overnight with primary antibodies (1:1,000) at 4˚C, followed by 2 h incubation with 1:1,000 dilution of secondary antibody for NLRP3 (anti-mouse IgG from GE Healthcare, Uppsala Sweden) and CARD8 (anti-rabbit IgG from AH Diagnostics, Stockholm, Sweden).

Statistical analysis

The independent two-tailed Student’s t-test and one way ANOVA was used for data analysis and results are expressed as mean ± SD. The P-value was considered to be statistically significant at P≤0.05.

Results

Effects of TNF-α on the expression and release of IL-1β and its regulators in AOSMCs

qRT-PCR revealed that stimulation of AOSMC with 50 ng/ml of TNF-α for different time points (1–48 h), induces a 2–3-fold increment of IL-1β mRNA expression, compared to their unstimulated controls (Fig. 1A). Furthermore, ELISA was used to quantify IL-1β protein release to the media from AOSMC stimulated with 50 ng/ml TNF-α and significant increases were observed after 24 and 48 h (Fig. 1B). We further analyzed the effects of TNF-α on IL-1β regulators: IL-1Ra, which is a natural competitor with IL-1β for the binding site to the IL-1 receptors; NLRP3, which is essential for NLRP3 inflammasome assembly; and CARD8 a co-inflammatory regulator. qRT-PCR revealed that stimulation of AOSMC with 50 ng/ml of TNF-α for different time points (1–48 h), induces a 1–3-fold increase in NLRP3 mRNA expression at time points 6, 24 and 48 h and a 2.5–6-fold increment of IL-1Ra mRNA expression at 24 and 48 h, compared to their respective unstimulated controls (Fig. 1C and D). CARD8 levels were not affected by TNF-α (data not shown).

Figure 1. Effects of 50 ng/ml TNF-α at different times points (1-48 h), on the expression and release of IL-1β and its regulators in AOSMCs compared to the unstimulated controls. (A) qRT-PCR expression of IL-1β mRNA, (B) ELISA quantification of IL-1β protein secreted, (C) qRT-PCR expression pattern of NLRP3 mRNA, (D) qRT-PCR expression pattern of IL-1Ra mRNA. (A, C and D) All values were normalized with cyclophilin A. *P≤0.05, **P<0.01 and ***P<0.001.

Effects of NLRP3 knockdown on IL-1β expression and release

Specific siRNAs were used to knockdown the NLRP3 gene to investigate the effect on the expression and release of IL-1β in AOSMC (Fig. 2A). The knockdown of NLRP3 was associated with a significant decrease in IL-1β expression and release (Fig. 2B and C), while a non-significant, but borderline increment of IL-1Ra mRNA expression was revealed (P=0.07; Fig. 2D). We also found that there was no difference in IL-1Ra protein release 24 h after NLRP3 knockdown, compared to the control (data not shown).

Figure 2. NLRP3 regulates IL-1β expression and secretion. AOSMC transfected with NLRP3 siRNA, followed by 24 h treatment with 50 ng/ml TNF-α, compared to the control without siRNA treatment. (A) Expression pattern of NLRP3 mRNA and protein. (B) Expression pattern of IL-1β mRNA. (C) IL-1β protein secreted. (D) Expression pattern of IL-1Ra mRNA. (A, B and D) In all experiments, values were normalized with cyclophilin A. n=5, *P≤0.05, **P<0.01 and ***P<0.001.

Effects of CARD8 knockdown on IL-1β expression and release

CARD8 has been reported to inhibit caspase-1 activity and also negatively regulate NF-κB activation in monocytes and other cell lines (15). We therefore hypothesized that knockdown of CARD8 may increase the expression and release of IL-1β. However, no significant change was evident in IL-1β mRNA (Fig. 3B) and protein release (Fig. 3C) but the IL-1Ra mRNA expression showed a non-significant, but borderline increment (P=0.08) (Fig. 3D) in AOSMC after knockdown of CARD8.

Figure 3. CARD8 regulates IL-1β expression and secretion. AOSMC were transfected with CARD8 siRNA, followed by 24 h treatment with 50 ng/ml TNF-α, compared to the control without siRNA treatment. (A) Expression pattern CARD8 mRNA and protein. (B) Expression pattern of IL-1β mRNA. (C) IL-1β protein secreted. (D) Expression pattern of IL-1Ra mRNA. (A, B and D) In all experiments, values were normalized with cyclophilin A. n=7, *P≤0.05, **P<0.01 and ***P<0.001.

Expression of CARD8 isoforms in AOSMCs

Considering the contradictory functional role reported between the CARD8 (T54 and T48) isoforms (19), we quest to determine which isoforms of CARD8 are expressed in AOSMC. We found that AOSMC express the T47, T48 and T54 isoforms and they were all knockdown by our siRNA. THP1 cells were used as positive control since it has previously been shown to express several CARD8 isoforms (19). Our results show that T54 is 250 bp, T48 is 200 bp while T47 is 300 bp (Fig. 4).

Figure 4. Comparative RT-PCR reveals that AOSMC express TUCAN 47, 48 and 54 isoforms (T47, T48 and T54, respectively). mRNA amplification of the three main isoforms was performed using the primer pair Tucan-54F/Tucan-Ex9-R for T54, Tucan-48F/Tucan-Ex9-R for T48 and Tucan-47F/Tucan-Ex9-R for T47. AOSMC were transfected with CARD8 siRNA, followed by 24 h treatment with 50 ng/ml TNF-α, compared to the control without siRNA treatment. THP1 cells were used as control.

Discussion

Existing evidence suggests that vascular cells actively contribute to atherogenic inflammation as they express several inflammatory moieties, which are overexpressed in atherosclerotic plaques. It is now evident that NLRP3 forms an integral part of the inflammasome that catalyzes the maturation of inactive IL-1β and IL-18 (24). CARD8 on the other hand has been reported to be an inhibitor of caspase-1 and a negative regulator of NF-κB and conflicting evidence has been shown regarding CARD8 and some inflammatory diseases like IBD and RA (15,17,18,25).

Consistent with other studies, we also found that TNF-α affects IL-1β expression and release (21,26). Expression of IL-1β and NLRP3 mRNA increased proportionally with time, while IL-1Ra significantly increased at 24 and 48 h after treatment. This suggests that, during cellular stress or exposure to a foreign ligand, elevated levels of IL-1β are required to execute its protective role, but over time excess production of IL-1β could become detrimental. Thus, the increase of IL-1Ra after 24 h could be to compensate for IL-1β activity and to maintain cellular homeostasis. In addition, our data supports the hypothesis that pathogens, cellular stress or cytokines (TNF-α) stimulate NLR’s to form inflammasome complexes which leads to the activation and secretion of IL-1β. To study the role of NLRP3 and CARD8 in IL-1β expression and release from AOSMC, cells were transfected with either siRNA against the NLRP3 or CARD8 genes. We observed that knockdown of NLRP3 led to a decrease of IL-1β mRNA and protein secretion. Our results are similar to findings reported by Zhu et al (27), who showed that silencing of the NLRP3 gene using siRNA, decreased the IL-1β release from Kupffer cells, hence antagonizing liver ischemic injury. The reason for the decreased IL-1β mRNA after NLRP3 knockdown is unclear, but we suggest that, IL-1β mRNA may undergo self regulation, due to the decrease in IL-1β protein secreted from the cells. Our results are consistent with studies carried out by Yamasaki et al (28), and Kankkunen et al (29). On the other hand, from the observation that CARD8 is an inhibitor of caspase-1 and a negative regulator of NF-κB (15), we derived a hypothesis that knockdown of the CARD8 gene may augment caspase-1 and NF-κB activities and thus increase the IL-1β expression and release. However, no significant change in IL-1β mRNA or protein release was evident.

Contrary to NLRP3, the role of CARD8 in IL-1β and IL-1Ra expression and release in AOSMCs may therefore be limited. However, a contradictory functional role has been demonstrated in the CARD8 isoforms as they show different expression pattern in different cell types and tumor cell lines with some cells expressing both (19). We further sought to determine which isoforms of CARD8 are expressed in AOSMC and found an expression of the 47, 48 and 54 kDa isoforms and that they were all knockdown by siRNA. However, due to lack of studies regarding CARD8, more studies are required to fully elucidate the functions of CARD8 and to demonstrate its role in inflammasome formation and activity.

In conclusion, our results show that TNF-α induces IL-1β, IL-1Ra and NLRP3 but not CARD8; knockdown of NLRP3 gene significantly decreases 1L-1β expression and release and AOSMC express the 47, 48 and 54 kDa isoforms of CARD8, but the effect of CARD8 on IL-1β and IL-1Ra seems to be limited. Our data therefore suggest that NLRP3 but not CARD8 play an important role in IL-1β expression and release in AOSMC and could be a therapeutic target or marker for atherosclerosis. Increased understanding of the mechanisms of initiation and progression of atherogenic vascular inflammation may lead to new approaches in the development of novel therapeutics for atherosclerosis and other inflammatory diseases.

Acknowledgements

This study was supported by grants from the Swedish Research Council, the Swedish Heart-Lung Foundation, the Swedish Fund for Research without Animal Experiments, the Swedish Heart and Lung Association, the Foundation of Olle Engkvist.

Abbreviations:

References