Expression of IL-1β, IL-1 Receptor Type I and IL-1 Receptor Antagonist in Human Aortic Smooth Muscle Cells: Effects of all-trans-Retinoic AcidWågsäter D.a · Jatta K.a · Ocaya P.a · Dimberg J.b · Sirsjö A.a
aDivision of Biomedicine, Department of Clinical Medicine, University of Örebro, Örebro, and bDepartment of Natural Science and Biomedicine, University College of Health Sciences, Jönköping, Sweden Corresponding Author
The proinflammatory cytokine interleukin (IL)-1β and the IL-1 receptor antagonist are expressed by atherosclerotic plaques and may be linked to the development of atherosclerosis. Existing evidence shows that retinoids and their receptors are involved in inflammatory response and that they are found in atherosclerotic plaques. In all-trans-retinoic acid (atRA)-treated human aortic smooth muscle cells (AOSMC), significant increases in IL-1β levels were observed, compared with untreated cells. Examination of IL-1 receptor antagonist and IL-1 receptor type I levels did not show any difference between atRA-treated and -untreated AOSMC. The results show that atRA-treated AOSMC express both the precursor (33 kDa) and the active form (17 kDa) of the IL-1β protein. atRA-treated carotid lesions showed significantly elevated IL-1β mRNA levels (2.9 ± 2.33) compared with untreated lesions (2.0 ± 1.77; p < 0.05). These results support the role of atRA as a regulator of inflammation such as in atherosclerosis.
Copyright © 2006 S. Karger AG, Basel
Interleukin (IL)-1β is an important mediator of the inflammatory response, is involved in cell proliferation, differentiation and apoptosis, and has been implicated in the development of atherosclerosis. The protein encoded by this gene is a member of the IL-1 cytokine family which consists of IL-1β, IL-1α and the IL-1 receptor antagonist (IL-1ra), located on human chromosome 2 . IL-1β is produced as a proprotein (33 kDa) which is proteolytically processed to its active form (17 kDa) partly by caspase 1 [2, 3]. The effects of the IL-1 gene family are directly correlated on their binding to two distinct surface receptors, IL-1 receptor type I (IL-1RI) and type II, whereas signal transducing ability only occurs after receptor type I binding [4,5,6,7,8,9].
Early findings by Moyer et al.  showed that endothelium and vascular smooth muscle cells (SMC), in conjunction with macrophages, serve as localized sources of IL-1 protein synthesis within the atherosclerotic lesion. These findings suggest that vascular cells may directly contribute to the pathogenesis of atherosclerotic vascular disease by actively secreting potent biologic mediators that modify vascular and immune cell function. Using atherosclerotic animal models, IL-1ra–/–/apoE–/– mice showed a 30% increase in lesion size, whereas IL-1β–/–/apoE–/– mice showed a 30% decrease compared with their wildtypes . These results suggest an important role of both IL-1ra and IL-1β in the modulation of lesion development [12,13,14,15].
Our group has previously reported the presence of retinoids and their receptors in atherosclerotic lesions . Retinoids have also been shown to modulate inflammation and immune response in a variety of cell types via nuclear receptors, the retinoic acid receptor (RAR) and the retinoid X receptors, as ligand-inducible transcription factors .
An earlier report has showed that all-trans-retinoic acid (atRA) induces the production and secretion of IL-1β and the downregulation of IL-1ra in human alveolar macrophages . The major objective of the present study is to determine the effect of atRA on the production and release of IL-1 from aortic SMC (AOSMC).
Material and Methods
Human AOSMC and their growth media were purchased from Cambrex BioScience (Walkersville, Md., USA) catalog No. CC-2571 with recommended supplements CC-3182 bullet kit. The medium was replaced every 3 days. For experiments, cells were grown to confluence and then incubated with or without atRA for 24, 48 and 72 h for analysis of mRNA and protein expression.
Carotid specimens from 8 patients were taken from plaques removed during carotid endarterectomy. All patients were operated on due to symptomatic carotid artery stenosis >70%. The arterial specimens were washed immediately in phosphate-buffered saline (PBS) to remove peripheral blood cells. All specimens contained advanced atherosclerotic lesions. Each arterial specimen was then divided into two sections, approximately 700 mg each. The tissue sections were incubated separately for 6 h at 37°C in 2 ml Dulbecco’s modified Eagle’s medium (DMEM/F12; Gibco, Rockville, Md., USA) containing 2% albumin (Biovitrum AB, Stockholm, Sweden) with DMSO or 1 μM atRA. Plaque tissue sections were then frozen in liquid nitrogen.
Total RNA from AOSMC was extracted with UltraspecTM (Biotecx laboratories, Houston, Tex., USA) as described by the manufacturer. To obtain RNA from the carotid lesions, the plaque tissues were first disrupted in a Mikro-Dismembrator II (B. Braun), and an RNeasy fibrous tissue mini kit (Qiagen, Valencia, Calif., USA) was later used to isolate RNA according to the manufacturer’s instructions. Recovered RNA was then stored at –70°C until use.
Total RNA (1 μg from AOSMC and 0.2 μg from carotid lesions) from each sample was reverse-transcribed to cDNA by using superscript II (Invitrogen, Carlsbad, Calif., USA) according to the manufacturer’s instructions. Real-time quantitative RT-PCR was performed on ABI Prism Sequence Detection System 7000 (PE Applied Biosystems, Foster City, Calif., USA). cDNA (3 μl) was amplified by real-time PCR with 1× TaqMan Universal PCR Master Mix (Roche, N.J., USA), 1 μM of both forward and reverse primers and 0.25 μM probe. Sequences for the amplification of IL-1β are previously described , whilst for the amplification of IL-1ra and IL-1RI, ‘assay on demand’ products were used (Applied Biosystems, catalogue numbers Hs00277299_m1 and Hs00168392_m1).
Gene expression was normalized to β-actin gene expression . Each sample was analyzed in duplicate under the following conditions: 2 min at 50°C, 10 min at 95°C, 0.15 min at 95°C and 1 min at 60°C. PCR amplification was correlated against a standard curve. Reactions were performed in MicroAmp optical 96-well reaction plates (PE Applied Biosystems).
Enzyme-linked immunosorbent assay (ELISA) was used for detection of IL-1β and IL-1ra protein in media from AOSMC. Medium was collected from cultured cells and centrifuged to remove cell debris. The samples were analyzed by ELISA on a Maxisorb® plate (Nunclon, Roskilde, Denmark) coated with mouse anti-human IL-1β or IL-1ra antibody according to the manufacturer’s recommendations (R&D Systems, UK). After washing in PBS-0.05% Tween, biotinylated goat anti-human IL-1β or IL-1ra antibody (R&D Systems) was used as detection antibody. Unbound antibody was washed off, and streptavidin-horse radish peroxidase (R&D Systems) was added. After washing in PBS-0.05% Tween, bound antibody was visualized with substrate solution (R&D Systems); 1 M H2SO4 was added to stop the reaction. The optical density was determined by measuring the absorbance at 450 nm. The absorbance was correlated against a standard curve.
Protein lysates (10 μg) from cells were separated under reducing conditions by electrophoresis using 10% SDS-PAGE as described previously . Separated proteins were transferred onto nitrocellulose membranes (Amersham, UK) in a transblot electrophoretic transfer cell (Bio-Rad Laboratories, USA). Filters were probed over night with a rabbit antihuman polyclonal IL-1RI antibody (Santa Cruz Biotechnology, Calif., USA), diluted 1:300, or with mouse antihuman monoclonal IL-1β (R&D Systems), diluted 1:700, in TTBS containing 3% (w/v) non-fat dry milk . Filters were developed using an enhanced chemiluminescence system (Amersham) and exposed to Hyperfilm enhanced chemiluminescence (Amersham).
Results are expressed as means ± SD. Differences of mRNA and protein levels were examined by a two-tailed Student’s t test. Differences of IL-1β mRNA levels from carotid lesions were examined by the Wilcoxon signed rank test. The results were considered significant at the level of p < 0.05.
Real-time RT-PCR was used to measure levels of IL-1β mRNA, which showed increased levels of about 2–3 fold after 24, 48 and 72 h of atRA-stimulated AOSMC, compared with unstimulated AOSMC (fig. 1a). ELISA was used to measure protein levels of IL-1β in media from AOSMC stimulated with 1 μM atRA for 24, 48 and 72 h. The results unveiled induced levels of IL-1β in media from AOSMC stimulated with atRA, compared with unstimulated cells. The induction, which increased over time, starting after 48 h, peaked with a 4-fold increase after 72 h (fig. 1b). Addition of 0.01–10 μM atRA to AOSMC induced a 1- to 3-fold increase in IL-1β expression after a 48-hour incubation (fig. 1c). However, significant induction was achieved with 0.1–10 μM atRA (fig. 1c). Western blot was used to examine the form of IL-1β expressed by AOSMC. The results showed that the precursor (33 kDa) and the active form of IL-1β (17 kDa) were expressed in cell lysates from AOSMC (fig. 1d) treated with atRA. Tumor necrosis factor-α-stimulated cells were used as a positive control (fig. 1d).
|Fig. 1. IL-1β is induced by atRA in AOSMC. a Expression of IL-1β mRNA in AOSMC, examined by real time RT-PCR. Data are shown as relative values to β-actin gene expression in means ± SD of four separate experiments. b Protein levels of IL-1β in media from AOSMC, examined by ELISA. c Concentration curve demonstrating a dose-dependent increase in IL-1β expression after 48 h incubation examined by ELISA. d Protein expression of the IL-1β precursor (33 kDa) and the active (17 kDa) form of IL-1β in lysates of AOSMC treated with atRA 48 h, examined by Western blot analysis. Tumor necrosis factor-α (TNF-α) is included as a reference. In all experiments, data are shown as means ± SD of four separate experiments with an exception in c, where one extremely high outlier (>701 pg/ml) was excluded from the 10 μM atRA-treated group. * p < 0.05; ** p < 0.01; *** p < 0.001.|
When IL-1RI mRNA levels were measured in AOSMC by real-time RT-PCR, over time, an increase in IL-1RI was observed (fig. 2a), and a similar increase was seen at protein levels analyzed by Western blot (fig. 2b). However, no difference in expressed levels of IL-1RI between atRA-treated and -untreated cells could be detected (fig. 2a, b).
|Fig. 2. Expression of IL-1RI by AOSMC is not dependent on atRA stimulation. a IL-1RI mRNA levels were measured in AOSMC by real-time RT-PCR. Data are shown as relative values to β-actin gene expression in means ± SD of four separate experiments. b Protein levels of IL-1RI in cell lysates from AOSMC cultured for 24, 48 and 72 h, examined by Western blot. The results showed no difference in expressed levels between atRA-treated and -untreated cells of IL-1RI over time.|
Since IL-1ra is the natural competitor to IL-1β of binding IL-1RI, we further analyzed the effects of atRA on expression IL-1ra in AOSMC. RT-PCR analysis of AOSMC treated with or without atRA showed that atRA had no effects on gene expression of IL-1ra in AOSMC (data not shown). We also performed ELISA analysis on media from atRA-treated and -untreated AOSMC. Differences in protein expressions were not seen (data not shown).
Carotid lesions from 8 patients were obtained during carotid endarterectomy. All patients were operated on due to symptomatic carotid artery stenosis >70%. Each lesion was divided into two equal parts and then incubated in serum-free conditioned media for 6 h at 37°C. Carotid lesions treated with 1 μM atRA showed significantly elevated IL-1β mRNA levels (2.9 ± 2.33) compared with untreated lesions (2.0 ± 1.77; p < 0.05) when examined by RT-PCR. Examination of IL-1ra mRNA and IL-1RI mRNA levels did not show any difference in the expression between atRA-treated or -untreated carotid lesions (data not shown).
This report shows the expressions of IL-1β, IL-1ra and IL-1RI in AOSMC, and that the gene expression of IL-1β mRNA is induced by atRA in AOSMC and in carotid lesions. These results provide additional information about atRA as an important regulator of inflammation, such as in atherosclerosis.
The finding in our paper that atRA induces IL-1β in AOSMC is consistent with other results in other cell types. For example in keratinocytes, atRA is able to induce IL-1β and enhances PMA-induced IL-1β in peripheral blood mononuclear cells [20, 21]. In untreated AOSMC, we could not detect IL-1β precursor (33 kDa) or active IL-1β (17 kDa). However, in atRA-treated AOSMC, both IL-1β precursor and active IL-1β were found [2, 3,22,23,24]. As shown in other studies, IL-1β precursor is the predominant form found intracellularly [25,26,27,28]. Dendritic cells accumulate IL-1β upon stimulation but do not secrete active IL-1β until in contact with CD8+ T cells which were shown to induce the release of active IL-1β via mechanisms other than through caspase-1 activation [28, 29]. Cultured media that contained atRA-treated AOSMC consisted mainly of the precursor form of IL-1β (data not shown). However, Hazuda et al.  showed that the IL-1β precursor can be cleaved extracellularly by novel proteases released at sites of inflammation, e.g. elastase, cathepsin G and collagenases.
In this study, we have shown that atRA significantly increases mRNA levels of IL-1β, but did not initiate any changes in the expression levels of IL-1ra or IL-1RI in human carotid lesions. However, we would like to stress that other cell types than AOSMC may account for the induction of IL-1β mRNA found in the lesions. For example, macrophages are well-known sources of IL-1β in atherosclerotic lesions. The IL-1β gene has a putative binding site for RAR [31, 32] making it possible that atRA-activated RAR can directly bind to this IL-1β gene enhancer region. Husmann et al.  showed in their study that the atRA response of the IL-1β promoter is ligand specific and depends on RAR. In a previous study, we detected RAR ligands in atherosclerotic lesions, ligands that can be formed by retinoldehydrogenases, which oxidize retinol to generate retinoic acid . Taken together, these results indicate that atRA may play a role in vivo in regulating IL-1β in carotid lesions .
In conclusion, the results of this study unveil that atRA induces IL-1β in AOSMC. Interestingly, carotid lesions treated with atRA were also found to express higher levels of IL-1β mRNA, compared with control lesions. The effects that atRA possess may result in an increased inflammatory response in atherogenesis. These results support the knowledge of atRA as a regulator of inflammation, such as in atherosclerosis.
This study was supported by grants from the Swedish Medical Research Council (K2002-71X-02042-36A), the Swedish Heart-Lung Foundation and the Swedish Health Care Sciences Postgraduate School (NFVO) Karolinska Institutet. We thank Fredrik Atterfelt for his excellent technical assistance.
Dr. Ken Jatta
Division of Biomedicine, Department of Clinical Medicine
University of Örebro
SE–701 82 Örebro (Sweden)
Tel. +46 73648 3877, Fax +46 1930 3778, E-Mail firstname.lastname@example.org
D.W. and K.J. contributed equally to this article.
Received: October 21, 2005
Accepted after revision: May 3, 2006
Published online: June 28, 2006
Number of Print Pages : 6
Number of Figures : 2, Number of Tables : 0, Number of References : 34
Journal of Vascular Research (Incorporating 'International Journal of Microcirculation')
Vol. 43, No. 4, Year 2006 (Cover Date: July 2006)
Journal Editor: Pohl, U. (Munich)
ISSN: 1018–1172 (print), 1423–0135 (Online)
For additional information: http://www.karger.com/JVR