Jump to
Abstract
Objective— Activation of transcription factor NF-κB is an important step in the development of vascular damage, because it controls inducible genes, including many inflammatory mediators. The pharmacological modulation of this process is the main objective in the design of new therapies for atherosclerosis. In this work we analyzed the effects of the natural compound parthenolide (PTN), an NF-κB inhibitor.
Methods and Results— In vascular smooth muscle cells (VSMCs) and monocytes stimulated with lipopolysaccharide (LPS), nontoxic doses of PTN reduced IκBα degradation, NF-κB activation, and MCP-1 expression, without inhibiting AP-1 and MAPK. In apoE mice, treatment with low (2 mg/kg, 20 weeks), medium (4 mg/kg, 10 weeks), and high (10 mg/kg, 10 weeks) dose of PTN reduced the size of aortic lesion, decreased macrophage, and increased VSMC content in the lesions. Treated mice showed reduced serum levels of MCP-1 and attenuated NF-κB activity, but not AP-1, in the lesions. Moreover, PTN affects neither apoptotic cell death nor oxidative stress in cultured cells and mice.
Conclusion— NF-κB inhibition by PTN retards atherosclerotic lesions in apoE mice, by reducing lesion size and changing plaque composition. This natural compound could represent a novel therapeutic approach to inflammation during vascular damage.
Atherosclerosis is a disease with a prominent inflammatory component, where monocyte recruitment and vascular smooth muscle cell (VSMC) proliferation and migration are essential for the initial stages. The progression of lesions is favored by lipid oxidation and inflammatory mediator release by macrophages, one promoting the other in a vicious cycle of events.1,2
The inflammatory gene expression is strictly regulated by transcription factors, such as nuclear factor-κB (NF-κB). The NF-κB family comprises five members (relA [p65], relB, c-Rel, p105/50, and p100/52) associated as homo or heterodimers.3 When stimulated, signaling through the upstream kinases (NIK and IKK) leads to phosphorylation of inhibitor IκBα, which is subsequently polyubiquitinated and degraded by the proteasome. Then, free NF-κB dimers translocate into the nucleus for target gene transcription activation.3,4 Although NF-κB activation is essential for normal physiological processes, its inappropriate and prolonged activation has been linked to inflammatory diseases, including rheumatoid arthritis and glomerulonephritis.5,6 Activated NF-κB has also been detected in mononuclear cells from atherosclerotic patients7 and in the lesions, mainly in VSMCs, macrophages, and endothelial cells.8
Major efforts have been focused on identifying novel antiinflammatory drugs, which can prevent the inflammatory process at the very early stage of gene expression. The natural product parthenolide (PTN), a sesquiterpene lactone isolated from extracts of Mexican Indian medicinal herb (Tanacetum parthenum), has attracted considerable interest.9 PTN-containing herbs have been commonly used for various inflammatory conditions such as migraine, arthritis, and asthma.10,11 Several groups have reported that PTN inhibits NF-κB activation in cultured cells12,13 and experimental models.14,15 Moreover, PTN exerts a strong proapoptotic activity on cancer cells,16 raising the potential use as an antitumor agent. However, its mechanism of action in atherosclerosis has not been investigated.
Because several lines of evidence support the idea that antiinflammatory interventions can inhibit the development of atherosclerosis and its clinical complications, in this study, we report the beneficial effect of PTN treatment in experimental atherosclerosis, partially attributed to the inhibition of NF-κB and subsequent expression of inflammatory genes.
Methods
Cell Cultures
Murine VSMCs,17 human VSMCs, and monocytes (CRL-1999 and THP-1 lines; ATCC, Rockville, MD) were cultured in medium containing 10% FCS (Life Technologies). Quiescent cells were treated for 90 minutes with PTN (Sigma Chemicals) then washed and stimulated with LPS (Sigma).
EMSA
Nuclear proteins (10 μg) were incubated with 0.035 pmol [32P]-NF-κB and -AP-1 oligonucleotides, and resolved on 4% nondenaturing polyacrylamide gels. Specificity was confirmed with 100-fold excess unlabeled probe.18
Fluorescence Analysis
Nuclear detection of NF-κB subunits was detected in fixed permeabilized cells with p50 and p65 Abs (Santa Cruz Biotechnology), followed by fluorescein isothiocyanate (FITC)-labeled Ab. Superoxide generation in cells and aortic sections was assessed using 5 μmol/L dihydroethidium (DHE; Molecular Probes),19 and number of positive cells was expressed as percentage of total cells (DAPI staining).
Western Blot
Cytosolic proteins (20 μg) were electrophoresed and transferred, and membranes were incubated with IκBα, followed by peroxidase-conjugated Ab and chemiluminescence.20
mRNA Expression
Total RNA was obtained with Trizol. MCP-1 expression was analyzed by RT-polymerase chain reaction (PCR) with mouse MCP-1 and GAPDH primers, and by Northern blot with human MCP-1 and 28S probes (JE/pGEM-hJE34 and HHCD07; ATCC). Relative amounts of MCP-1 were established in relation to GAPDH or 28S.20
Cell Proliferation and Apoptosis
Cell proliferation after 24 hour incubation in medium containing 10% FCS was determined in 96-well plates by colorimetric assay of methylene blue incorporation.20 Hypodiploid apoptotic cells after 24 hours in serum-free medium were quantified by flow cytometry of permeabilized, propidium iodide–stained cells.21
ELISA
Serum MCP-1 was measured using BD OptEIA Set (BD Pharmagen). MAPK activation was measured by cellular ELISA. In brief, 1×105 cells were treated with PTN, PD98059, or SB203580 (Calbiochem) before LPS stimulation. Fixed cells were incubated with phosphospecific Abs to extracellular signal regulated kinase (ERK), p38, and JNK (Santa Cruz) followed by peroxidase-conjugated Abs and color development. The 450 nm absorbance was normalized to total MAPK content and cell number.
Experimental Atherosclerosis
Male apoE mice (12 weeks of age; Jackson Laboratory, Bar Harbor, Me) fed on a Western diet were divided into two groups: PTN treatment (n=21) and control (n=12). The treated group was subdivided into low (2 mg/kg, 20 weeks; n=4), medium (4 mg/kg, 10 weeks; n=9), and high (10 mg/kg, 10 weeks; n=8) dose of PTN, which was ip injected every 2 days from the beginning of diet until sacrifice. Controls consisted on apoE mice injected with vehicle (10 weeks, n=6; 20 weeks, n=6) and wild-type (WT) receiving PTN or vehicle (n=4). All studies were performed in accordance to the EU normative. Cholesterol, triglycerides, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) were tested in serum samples from fasted mice.
Histology
Anesthetized mice were saline-perfused. Liver, spleen, and kidney were paraffin-included for histology. Aorta was OCT-embedded and sections with a 160 μm interval were Oil-Red-O/hematoxylin stained.22 After image analysis, individual lesion area was determined by averaging the maximal values (3 to 4 sections). Some sections were Masson trichrome stained for fibrosis, cellularity (score 0 to 4), and necrosis area determination. Macrophages, VSMCs, and apoptosis were detected by immunoperoxidase with Moma-2, α-actin, and ssDNA Ab,23 respectively. Activated NF-κB and AP-1 were detected by Southwestern histochemistry with digoxigenin-labeled probes, using competition and mutant probe as specificity controls.20,24 Colocalization was made by subsequent staining with Moma-2 or α-actin Abs. Data of computer-assisted morphometric analysis were expressed as percentage of positive staining per mm2 (immunohistochemistry) and positive nuclear staining per mm2 (Southwestern).7
Statistical Analysis
Data expressed as mean±SD were analyzed by ANOVA and Tukey-Kramer tests (P<0.05 was considered significant).
Results
PTN Inhibits NF-κB Activation in Cultured Cells
Murine VSMCs were treated with PTN 90 minutes before LPS stimulation, and NF-κB activity in nuclear extracts was analyzed by EMSA. PTN dose-dependently reduced the intensity of specific bands (% inhibition: 76±16; Figure 1A). Nuclear extract incubation with PTN did not decrease the band pattern, excluding an effect on DNA binding ability (not shown). To confirm that NF-κB pathway is also well inhibited in a human system, we analyzed the PTN effects on human VSMCs and monocytes. PTN blocked NF-κB activity in both cell types (Figure 1A and 1B), without affecting other transcription factors, such as AP-1. PTN also prevented the IκBα degradation and p65 nuclear translocation induced by LPS in VSMCs (Figure 1C and 1D). Additionally, PTN reduced the expression of MCP-1, a NF-κB-regulated gene, both in VSMCs (Figure 1E) and monocytes (n-fold versus basal 6 hours: LPS, 4.5±0.8; PTN 1 μg/mL, 2.6±0.6; PTN 3 μg/mL, 1.5±0.4; P<0.05; n=2).
Figure 1. PTN inhibits NF-κB activation and MCP-1 expression. Cells were preincubated for 90 minutes with different doses of PTN (μg/mL), then washed and stimulated with 1 μg/mL LPS. NF-κB and AP-1 binding activity in nuclear extracts from VSMC (A) and monocytes (B) at 30 minutes stimulation (com, competition assay). C, Western blot for IκBα in murine VSMC. D, Immunofluorescence for p65 subunit in human VSMCs. Representative gels and micrographs (200× magnification) from 2 to 5 experiments. E, RT-PCR analysis for MCP-1 in murine VSMCs. Fold changes (relative to basal) of mRNA levels are mean±SD of 4 experiments (*P<0.05 vs LPS).
We analyzed whether PTN modulates other signaling pathways, such as MAPK. In VSMCs, the maximal activation of ERK, p38, and JNK was not affected by increasing concentrations of PTN, as assessed by cellular ELISA (Figure 2A). Nevertheless, p38 and ERK phosphorylation was prevented by the specific inhibitors SB203580 and PD98059 (% inhibition: 97±3, and 89±4; n=3). Similarly, ERK, p38, and JNK activation in monocytes was not affected by PTN (LPS: 3.7±0.5, 3.5±0.4, and 3.3±0.1; PTN 3 μg/mL: 3.8±0.4, 3.3±0.5, 3.1±0.2, respectively; n-fold at 60 minutes, n=2).
Figure 2. MAPK activation, proliferation, apoptosis, and oxidative stress. A, Effect of PTN on the maximal MAPK activation by LPS in murine VSMC (ERK, 60 minutes; p38 and JNK, 120 minutes), determined by cellular ELISA. Proliferation (B) and apoptosis (C) at 24 hours in murine VSMC and THP-1 were assessed by methylene blue and flow cytometry, respectively. D, Superoxide production at 30 minutes in VSMCs was measured by DHE fluorescence. Bars are mean±SD of 3 to 4 experiments in triplicate (*P<0.05 vs basal).
PTN Effects on Cell Growth and Apoptosis
PTN was not cytotoxic against VSMCs and monocytes up to 10 μg/mL at all time-points measured (95% viability by trypan blue). In proliferation assays, PTN (1 to 10 μg/mL, 90 minutes) did not diminish the number of VSMCs and monocytes (Figure 2B). Decreased cell proliferation was evident after 24 hours of continuous exposure to higher concentrations (20 μg/mL, % versus control: VSMCs, 75±4; THP-1, 70±6; n=4). Apoptosis of VSMCs after 24 hours incubation in serum-free medium was not significantly impaired by PTN, as assessed with propidium iodide (Figure 2C). Interestingly, THP-1 cell death was moderately induced at higher concentrations (Figure 2C), in agreement with previous reports describing that PTN selectively prompt apoptosis in tumor but not in normal cells.16 Because apoptosis may be an oxidative stress–mediated process,21,25 we investigated the intracellular superoxide production in murine VSMCs by DHE fluorescence. PTN did not modify the number of positive cells both in basal conditions and after LPS stimulation (Figure 2D).
PTN Treatment in Experimental Atherosclerosis
We next analyzed the effect of PTN on atherosclerosis. Compared with WT, apoE mice fed on high-fat diet showed structural alterations of the vessel wall, accumulation of lipids within intima of the aortic root, and lesions covered a 55% of the cross-sectioned area (Figure 3A through 3C). PTN at low dose (2 mg/kg) moderately reduced the aortic lesion, whereas shorter treatment with higher doses (4, 10 mg/kg, 10 weeks) was more effective (Figure 3D through 3F). In any case, PTN significantly reduced both the size and extent of the lesions when compared with vehicle-treated controls (Figure 3G and 3H). Lesion quantification after Masson staining (Figure 4A and 4B) revealed that fibrosis and cellularity index decreased by PTN treatment (% inhibition: 47±8 and 45±11, respectively; P<0.01 versus untreated). Interestingly, PTN-treated mice displayed significantly higher VSMC content, with no changes in necrosis area, apoptotic cell number, and oxidative stress in the lesions (Figure 4E).
Figure 3. PTN treatment improves atherosclerosis. Oil-Red-O/hematoxylin staining in serial aortic sections from WT (A) and apoE mice fed with hyperlipidemic diet for 10 weeks (B, E, and F) and 20 weeks (C and D). Treatment with low (D), medium (E), and high (F) dose of PTN reduced lipid accumulation and lesion size. Magnification ×40. G, Lesion area throughout the studied region in untreated (▴) and 10 mg/kg PTN-treated (▪) mice (*P<0.01 vs untreated). H, Average of maximal lesion area for each group.
Figure 4. Changes in morphology and plaque composition by PTN treatment. Representative Masson trichrome staining (A and B) and macrophage immunohistochemistry (C and D) in sections from control (A and C) and 10 mg/kg PTN-treated (B and D) mice. Magnification ×200. Atherome (a), media (m), and lumen (l) are indicated. E, Quantitative analysis of necrosis area (Masson staining), apoptotic cells (ssDNA positive cells), superoxide production (DHE fluorescence), VSMC content (α-actin staining), and macrophages (Moma-2 staining) in the different groups. Data are expressed as mean±SD (*P<0.01 vs untreated).
Systemic effects of PTN were evaluated in liver, kidney, and spleen from apoE and WT mice injected with 10 mg/kg PTN. Hepatic morphology and cytology were normal, without signs of inflammation, necrosis, and hepatocyte alterations. Inflammation, mesangiolysis, and fibrosis were absent in renal tissues. Splenic architecture remains intact. Moreover, no significant differences in serum AST and ALT activities, liver damage indicators, were detected in all the studied groups. Serum levels of cholesterol and triglycerides were also unmodified by PTN treatment (supplemental Table I, available online at http://atvb.ahajournals.org).
PTN Reduces Inflammatory Responses in Atherosclerotic Mice
Because monocyte influx into the artery wall and further differentiation to macrophages is a key step in atherogenesis,1 we checked whether PTN regulates monocyte infiltration. By immunohistochemistry, the high number of macrophages located in the neointima of apoE mice was reduced by PTN treatment (Figure 4C through 4E). According to this, lower levels of the monocyte chemokine MCP-126 were found in sera from PTN-treated mice (pg/mL: untreated, 1579±39; PTN 4 mg/kg, 416±132, P<0.01).
The in vivo effects of PTN on transcription factor activation were evaluated by Southwestern histochemistry. Untreated mice showed strong nuclear staining for NF-κB (Figure 5A and 5B) colocalized with macrophages in the neointima and VSMCs in the media (Figure 5G and 5H). Nuclear staining was not observed with mutant probe (Figure 5I) or unlabeled consensus competition (not shown). The number of NF-κB–positive cells decreased in aortas from PTN-treated mice (nuclei/mm2: apoE, 5848±330; PTN 4 mg/kg, 4157±316; PTN 10 mg/kg, 3162±303; P<0.01; Figure 5D and 5E). By contrast, AP-1 was not significantly inhibited by PTN (nuclei/mm2: 4415±83; 3996±209; 3706±343; Figure 5C and 5F), confirming its specificity for NF-κB pathway.
Figure 5. NF-κB attenuation in PTN-treated mice. Representative Southwestern histochemistry for NF-κB (A and B) and AP-1 (C) in untreated mice showing strong nuclear staining. High dose PTN treatment inhibited NF-κB (D and E) but not AP-1 (F). Colocalization of NF-κB with macrophages (G) and VSMCs (H) in apoE sample. Arrows indicate double staining (blue nucleus, brown cytoplasm). I, Mutant NF-κB probe (same sample as G) as specificity control. Magnification (A and D) ×40, and details (B, C, and E through I) ×200. Atherome (a), media (m), and lumen (l) are indicated.
Discussion
NF-κB is one of the first mediators activated in the inflammatory response that plays a major role in atherogenesis, and its activation has been reported in atherosclerotic plaques and in cultured cells.7,27,28 In this work, we used the natural compound PTN in the treatment of experimental atherosclerosis. In vitro experiments with LPS-stimulated VSMCs and monocytes proved that PTN, at noncytotoxic doses, inhibited IκBα degradation, then preventing NF-κB activation and the subsequent MCP-1 gene expression. This is in agreement with previous studies on the regulation of interleukin (IL)-8, cycloxygenase, and iNOS by PTN in several cell types.20,29–31 We also report the beneficial effects of low, medium, and high dose of PTN on apoE mice fed on hyperlipidemic diet. Atheroma in apoE mice is more inflammatory than in humans, and this model better represents the initial stages than the complicated vulnerable plaques.32 The lesion area was reduced in all PTN-treated mice without modifying serum lipid levels, indicating no direct effects on the lipid metabolic pathway.
Regulation and control of NF-κB can be achieved by gene modification therapies or by pharmacological inhibition of key components of the cascade.4 Several common drugs, such as statin or aspirin, exert their beneficial effects on atherosclerosis by inhibiting different routes of cell signaling, including NF-κB.17,33 More recently, the natural compounds magnolol and flavopiridol have emerged as potential NF-κB modulators,34,35 although their precise atheroprotective mechanisms have not been well described. Other approaches, such as NF-κB decoy oligodeoxynucleotides and mutated IκBα overexpression, have also been successfully applied in animal models of vascular disease.36,37
Although atherosclerosis development in apoE mice is known to be associated with a genetic deficiency in lipid metabolism, evidence indicates that immune and inflammatory responses may be directly related to metabolic anomalies.38 Several groups suggest a link between inflammation and plaque rupture in thinner regions, because a large amount of infiltrating cells, principally monocytes, are found in this vulnerable region. In our study, PTN treatment reduced both the number of monocytes into the lesion area and the serum levels of MCP-1, a chemokine found in human plaques39 which overexpression increases lesion progression in mice.40 Moreover, the high relative ratio of VSMCs/macrophage in the lesions of PTN-treated animals is suggestive of a more stable plaque phenotype.
PTN effectiveness was also evident in NF-κB activation during atherosclerosis, with a reduced number of activated cells in the intima and media of PTN-treated mice. This is in accordance with previous articles describing the beneficial effects of PTN on myocardial ischemia, endotoxic shock, and renal disease through NF-κB inhibition.15,20 Although the molecular mechanism of NF-κB inhibition by PTN has been recently studied, it is still a matter of debate. Evidence indicates that PTN may inhibit IκBα phosphorylation and degradation,15,20,41 directly affect the DNA-binding ability without changing IκBα degradation,14 or block signaling pathways other than NF-κB.21,42 We found that PTN prevented NF-κB activation and dependent gene expression in human and murine cells, without affecting AP-1, consistent with previous articles.20,43 It has also been reported that PTN greatly sensitized cancer cells to tumor necrosis factor (TNF)-induced apoptosis, by NF-κB and p38 suppression, and sustained JNK activation.44 However, we did not observe such inhibitory effect of PTN on p38, ERK, and JNK in VSMCs and monocytes, which might be due to the differences of cell lineages used or the experimental procedures applied. Then, we conclude that the effects on atherosclerosis are mainly mediated through NF-κB inhibition. Several authors have described that PTN exerts its cytotoxic effects in a tumor cell–specific manner, without affecting control cells, and that both cell growth inhibition and apoptosis contribute to these anticancer effects.16,45 Recently, Li-Weber et al pointed out the dual role of PTN in regulating life and death of cells. In lymphocytes, PTN may function as antioxidant or cause oxidative stress–mediated apoptosis.25 In our work, cell viability and proliferation were unchanged by PTN, at the doses and incubation times that effectively inhibited NF-κB activation and target gene expression. Moreover, oxidative stress, apoptosis, and necrosis did not increase in aorta from PTN-treated mice, suggesting that PTN by itself does not induce cell death in the atherosclerotic lesions.
To our knowledge, this is the first study showing that PTN protects against atherosclerosis, an effect mainly mediated by NF-κB inhibition. However, the benefit of inhibiting NF-κB pathway in atherosclerosis is a current controversy. Some authors propose that NF-κB could act as a proinflammatory regulator of the early inflammation, and as antiinflammatory regulator of the resolution of inflammation.46 Ex vivo studies with cells from human atherosclerosis plaques revealed that canonical pathway of NF-κB activation selectively regulate proinflammatory and prothrombotic mediators expression in this disease.47 Interestingly, p50-deficient cells transplantation decreased lesion in atherosclerotic mice, but this lesion was characterized by an inflammatory phenotype.48 Although PTN therapy reduces inflammation in apoE mice without adverse symptoms, we cannot discard that longer treatment might cause unexpected side effects such as immunosuppression or deteriorated host defense. Because NF-κB is a major determinant for immunologic and inflammatory responses, further studies are needed to clarify the protective or pathological role of NF-κB in human atherosclerosis.
In summary, our results indicate that PTN prevents NF-κB activation in cells present in atherosclerotic plaques, and retards atherosclerosis progression, by reducing the extension and size of the lesions, and the inflammatory cell content. This natural compound could represent a novel therapeutic approach to inflammation during vascular damage.
Acknowledgments
We thank Drs L. Blanco-Colio, J.L. Martin-Ventura, B. Santamaria, B. Cano, and A. Kuhn for technical assistance and helpful comments.
Sources of Funding
This work was supported by grants from FIS (PI02/0539), CAM (08.4/0014/2001), MEC (2005/05857), and EU (QLRT-2001-01215).
Disclosures
None.
Footnotes
-
Original received April 18, 2005; final version accepted May 19, 2006.
References
This Issue
Jump to
Article Tools
Share this Article
- Parthenolide Modulates the NF-κB–Mediated Inflammatory Responses in Experimental Atherosclerosis
