Cigarette smoke is associated with increased carotid intimal thickening or stroke. Preliminary work showed that exposure to smoke resulted in a 4.5-fold reduction of heat shock protein-70 (HSP70) expression in spleens of mice using gene microarray analysis. In the current study, we investigated the role of extracellular HSP70 in carotid intimal thickening of mice exposed to cigarette smoke. Intimal thickening was induced by placement of a cuff around the right carotid artery of mice. Cuff injury resulted in increased HSP70 mRNA expression in carotid arteries that persisted for 21 days. Cigarette smoke exposure decreased arterial HSP70 expression and significantly increased intimal thickening compared with mice exposed to air. Treatment of mice exposed to cigarette smoke with intravenous recombinant HSP70 attenuated intimal thickening through reduced phosphorylated extracellular signal-regulated kinase (pERK) expression in the arterial wall. In vitro experiments with rat aortic smooth muscle cells confirmed that recombinant HSP70 decreases pERK and proliferating cell nuclear antigen (PCNA) expression in cells exposed to cigarette smoke extract and H2O2. Our study suggests that decreased expression of arterial HSP70 is an important mechanism by which exposure to cigarette smoke augments intimal thickening. The effects of recombinant HSP70 suggest a role for extracellular HSP70.
- arterial injury
- oxidative stress
- extracellular signal-regulated kinase
cigarette smoking is associated with increased carotid intimal thickening or incidence of stroke (4, 10, 17, 20, 29). We have previously reported that exposure to cigarette smoke accelerates the development of arterial intimal thickening in mice after injury (1, 35). However, the underlying mechanisms of these deleterious effects are unclear. We performed preliminary studies with gene microarray analysis of spleens from mice exposed to cigarette smoke and found a 4.5-fold decreased expression of stress-response heat shock protein-70 (HSP70) compared with mice exposed to air, identifying HSP70 as a potential target for further mechanistic studies.
HSPs are part of the host stress-response mechanism. Arterial injury results in oxidative stress (1, 6, 14), and evidence suggests that the oxidative stress response in atherosclerosis involves HSPs (31, 36). HSP70 is overexpressed in atherosclerotic lesions (16). Higher serum levels of HSP70 are associated with reduced atherosclerotic intimal thickening and lower risk of coronary artery disease (38). Furthermore, thermal treatment of rats increased arterial wall HSP70 expression and attenuated intimal thickening after injury (27).
HSP70 modulation of cell signaling is, in part, mediated by sequestration of the Raf-1-activator, Bag-1, and decreased phosphorylated extracellular signal regulated kinase (pERK) activation (33). These roles of HSP70 have been attributed to its intracellular functions. However, recent studies also point to a role of HSP70 in the extracellular milieu and provide evidence of specific pathways for stressed cells to actively release extracellular HSP70 (2).
In this study, we investigated the role of HSP70 in intimal thickening augmented by cigarette smoke exposure in a periadventitial carotid cuff injury model. We confirmed that exposure to cigarette smoke decreases HSP70 in the arterial wall leading to augmented intimal thickening after cuff placement. We further show that extracellular HSP70 modulates the detrimental effect of exposure to cigarette smoke on intimal thickening.
MATERIALS AND METHODS
Male wild-type C57BL/6J mice were purchased from Jackson Laboratories at 6 wk of age and were fed regular chow throughout the duration of the experiment. At 20 wk of age, the mice were exposed to cigarette smoke (standardized reference research cigarettes 1R1 cigarette from University of Kentucky Tobacco and Health Research Institute) in a specially designed apparatus (Department of Physiology, Washington University School of Medicine). The smoking apparatus consists of five chambers each connected through a plastic tube to a syringe. The syringe aspirates cigarette smoke from the cigarette placed in a special holder and delivers it to a chamber that holds the mice. The mice were exposed to cigarette smoke starting from 1 puff at the age of 20 wk. The number of puffs was gradually increased each day until mice were routinely exposed to smoke from 1 cigarette per day. The mice were exposed to room air in an identical apparatus as control (1, 35). At the age of 25 wk, the mice were anesthetized with Avertin (0.017 ml/g of a 2% solution), and a nonocclusive plastic cuff (length 2 mm; internal diameter, 0.51 mm; Cole-Parmer Instrument Co.) was aseptically placed around the right carotid artery and the skin incision was closed (1). Contralateral arteries served as uninjured control.
Mice were euthanized 21 days after cuff placement, and the carotid arteries were perfused with 0.9% saline for 10 min. Harvested injured carotid arteries were embedded in OCT compound (Tissue-Tek, Allegiance) and frozen at −80°C. Aortas and spleens were also harvested and snap-frozen in liquid nitrogen. Some aortic tissues were collected fresh for ex-vivo experiments (below). Blood was collected by retro-orbital bleed prior to euthanasia. Time points for the study of mRNA in carotid arterial tissues were as follows: no injury, 1 day, 7 days, and 21 days after cuff placement. Five carotid arteries were pooled for each time point to obtain sufficient quantity of RNA for analysis, whereas sufficient tissue was available from half of a single aorta or spleen.
Treatment with recombinant HSP70 (rHSP70).
Separate groups of mice exposed to cigarette smoke or air were given intravenous injections of recombinant mouse HSP70 protein (rHSP70-low endotoxin, Stressgen) at a dose of 10 μg/mouse in 0.9% NaCl (19, 34) with a final volume of 100 μl on the day of the injury, and 3, 7, 11, and 15 days after injury. Controls were treated with an equal volume of 0.9% NaCl.
The Institutional Animal Care and Use Committee approved the experimental protocols used in this study.
Ex vivo heat exposure.
Aortic tissues from mice euthanized 21 days after cuff placement were placed in 1% FBS-DMEM/F12 and heated ex vivo in 43°C for 1 h and transferred to 37°C; total RNA was extracted 3 h later.
Frozen sections (6 to 8 μm thick) were collected from the injured carotid arteries. Four sections were collected on each slide, and 20 to 25 slides were collected from each injured arterial segment, as described previously (6). Slides were stained with hematoxylin and eosin, and the vessel area measured using computer-assisted morphometric analysis (Image-Pro Plus). The measurements of sections from each animal were averaged for analysis.
Total RNA was extracted from spleens and carotid arteries using a commercially available kit (TRIzol; Invitrogen). One-microgram aliquots were subjected to reverse transcription using random primers and the Superscript II system (Invitrogen). Aliquots were then used for PCR using previously described primers for murine β-actin (23) and HSP70 (9). Cycling conditions were optimized for each primer set, and PCR products were run on 1.5% agarose gel stained with ethidium bromide. Gels were digitally photographed, and densitometric analysis was performed. Values were expressed as a ratio to β-actin.
Injured carotid tissues, aortic tissues, or aortic smooth muscle cells were homogenized in cold lysis buffer (10 mM HEPES, 1 mM EDTA, 60 mM KCl, 0.4% NP40) with proteinase inhibitor cocktail (Roche). After spinning at 12,000 g for 10 min, the supernatant was collected as cytosolic protein extract.
Western blot analysis.
Equal amounts of the extracted cytosolic protein were electrophoresed on 10% or 12% SDS-PAGE gel and transferred to nitrocellulose membrane. The membranes were incubated with the following primary antibodies at 4°C overnight: polyclonal antibody against HSP70 (1:200, Stressgen), Bag-1, ERK, p21 (1:200), PCNA (1:1,000; Santa Cruz Biotechnology) and phosphorylated ERK1/2 (1:200; Cell Signaling Technology). Detection was performed with horseradish peroxidase conjugated anti-goat or anti-rabbit antibody, and enhanced chemiluminescence (Amersham). The β-actin antibody (Santa Cruz Biotechnology) used was conjugated to horseradish peroxidase.
The sections from the injured carotid arteries were incubated with polyclonal HSP70 antibody (1:50, Santa Cruz Biotechnology), MOMA-2 (Serotec), and active caspase 3 (BD Bioscience). Biotinylated secondary antibody was used with the 3-amino-qethylcarbazole chromogen detection kit (DAKO Corp). Omission of primary antibody was used as a control. The semiquantitative measurement of the HSP70 stained area standardized against the medial area was done with computer-assisted analysis (7).
Detection of anti-HSP70 antibodies.
Relative serum HSP70 antibody levels were determined by ELISA. Briefly, 96-well ELISA plates (NUNC, Maxisorp) were coated with recombinant mouse HSP70 protein in PBS (2 μg/ml) overnight at 4°C. Diluted sera (1:100 in 1% BSA in PBS) were incubated on the washed plate for 1 h, washed, and detected with horseradish peroxidase-conjugated IgG antibody for 1 h. The substrate for color development was 2,2′-azino-bis-(3-benzthiazoline-6-sulfonic acid). Optical density was recorded 10 min later on a SpectraMax 190 reader (Molecular Devices).
In vitro experiments.
Rat aortic smooth muscle cells (RASMC) were grown in 10% FBS-DMEM/F12 to confluence. After incubating in 0.1% BSA-DMEM/F12 for 48 h, cells were treated with 200 μM H2O2 for 1 h, 3 h, and 24 h. Another set of experiments included pulse treatment with 25 μg/ml (1) cigarette smoke condensate (University of Kentucky Tobacco and Health Research Institute) for 1 h. Media were then replaced with fresh 0.1% BSA-DMEM/F12 medium containing 200 μM H2O2 and incubated in 37°C. For HSP70 treatment, exogenous recombinant HSP70 was added the same time as the H2O2 at 50 and 100 ng/ml concentrations. Cell cycling analysis was performed after harvested cells were fixed in ethanol and stained with propidium iodide. All experiments were performed within 15 passages.
Numeric data are expressed as means ± SD. Analysis was perfomed using ANOVA followed by multiple comparison using Neuman-Keuls test. Student's t-test was used for comparisons of two groups. A P < 0.05 was considered significant.
Effect of exposure to cigarette smoke on HSP70 expression.
Aortic HSP70 protein expression was significantly decreased in mice exposed to cigarette smoke compared with mice exposed to air (1.21 ± 0.55 vs. 7.40 ± 3.13 densitometric units, respectively, n = 3 each; P < 0.05; Fig. 1A). The effect of exposure to cigarette smoke appeared to be systemic as splenic HSP70 mRNA expression was also significantly decreased (Fig. 1B; P < 0.05). Ex vivo heat treatment of aortic tissue resulted in increased HSP70 mRNA expression in aorta of control mice. HSP70 mRNA expression in heat-treated aorta of mice exposed to cigarette smoke was decreased compared with mice exposed to air (Fig. 1C).
Arterial injury, HSP70, and effect of cigarette smoke.
To determine whether cuff injury would result in increased stress response, we assessed HSP70 mRNA expression at specific time points. Densitometric analysis indicated a 2-fold increased HSP70 mRNA expression in the carotid artery within a day after injury and persisted for 21 days (Fig. 2A).
The increase in HSP70 mRNA expression in arteries 1 day after injury was attenuated by 35% in mice exposed to cigarette smoke compared with mice exposed to air (Fig. 2B), as measured by densitometric analysis. The presence of HSP70 21 days after cuff placement was confirmed by immunostaining in the media and intima of the injured arteries. The percentage of HSP70-positive stain area was significantly decreased in mice exposed to cigarette smoke compared with mice exposed to air (5.1 ± 2.3% vs. 10.8 ± 4.9%, respectively; n = 4 each; P < 0.05; Fig. 2, C and D).
Intimal thickening measured 21 days after cuff placement was significantly increased in mice exposed to cigarette smoke compared with mice exposed to air (0.026 ± 0.01 mm2, n = 7 vs. 0.012 ± 0.013 mm2, n = 6; P < 0.05; Fig. 2, E and F). The intima/media ratio (I/M ratio) was also significantly increased in mice exposed to cigarette smoke compared with mice exposed to air (0.62 ± 0.23 vs. 0.31 ± 0.22; P < 0.05). No significant difference was observed in medial area, total vessel area, and MOMA-2 stain area (not shown).
Effect of exogenous rHSP70.
The effect of rHSP-70 treatment was tested in mice exposed to cigarette smoke. Mice were administered intravenous injections of rHSP70 at a dose of 10 μg/mouse in 0.9% NaCl (34) with a final volume of 100 μl on the day of the injury, and 3, 7, 11, and 15 days after injury. Controls were treated with an equal volume of 0.9% NaCl. At 21 days after cuff placement, HSP70 was detected in the intima and the adventitial side of the media of arteries from mice exposed to cigarette smoke and treated with rHSP70 (Fig. 2G). Treatment of mice exposed to cigarette smoke with rHSP70 resulted in significantly decreased intimal thickening compared with treatment with saline (Table 1; P < 0.01; Fig. 3, A and B). I/M ratio was also significantly decreased in mice treated with rHSP70 compared with mice treated with saline (Table 1; P < 0.01). Macrophage staining and active caspase-3 staining were not affected by rHSP70 treatment (not shown). In the mice exposed to air, treatment with rHSP70 had no significant effect on intimal thickening compared with no treatment (Table 1). Macrophage staining and active caspase-3 staining were also not affected by rHSP70 treatment in mice exposed to air (not shown).
To investigate the potential mechanism of the effects of rHSP70, ERK, and Bag-1 expression were assessed by Western blot analysis. Treatment with rHSP70 of mice exposed to cigarette smoke resulted in a reduction of pERK expression in injured carotid arteries 7 days after cuff placement compared with treatment with saline (0.75 vs. 0.29 densitometric units, respectively) but did not affect Bag-1 expression (Fig. 3C). Treatment with rHSP70 of mice exposed to air had no effect on pERK expression (Fig. 3D).
At 7 days after cuff placement, there were no differences in serum levels of HSP70 antibody in mice exposed to cigarette smoke and treated with rHSP70 compared with mice treated with saline (not shown).
Oxidative stress in vitro and HSP70.
We have previously reported that arterial injury using the cuff method results in increased oxidative stress (1). In vitro experiments were performed to define the signaling pathway affected by oxidative stress and cigarette smoke exposure on rat aortic smooth muscle cells. Treatment of confluent, 48-h serum-starved RASMC with 200 μM H2O2 resulted in increased HSP-70 expression after 24 h (Fig. 4A). β-actin was used as a loading control. Activated ERK (pERK-p44/p42) was increased within 1 h of H2O2 treatment. One-hour pulse treatment of RASMC with cigarette smoke extract (CSE) prior to the 24 h H2O2 treatment decreased HSP70 expression (Fig. 4B). However, pulse treatment with CSE prior to the 1 h H2O2 treatment augmented pERK expression compared with H2O2 alone (Fig. 4C). Treatment with CSE alone without H2O2 did not significantly affect pERK and HSP70 expression.
Cotreatment of CSE-pulsed and H2O2-treated RASMC with rHSP70 at doses of 50 and 100 ng/ml for 3 h decreased pERK and PCNA expression (Fig. 5, A–C). Cell cycle analysis showed that percentage of cells in S-phase after 48 h CSE and H2O2 treatment significantly increased compared with untreated cells (P < 0.01; n = 3, Fig. 5D). Cotreatment of CSE-pulsed and H2O2-treated RASMC with rHSP70 at doses of 50 and 100 ng/ml reduced the percentage of cells in S-phase (P < 0.05; n = 3, Fig. 5D). This suggests that rHSP70 treatment decreased activation and proliferation of RASMC stimulated with CSE and H2O2. PCNA expression was not affected by CSE treatment without H2O2 (not shown). p21 expression was significantly increased by cotreatment with CSE pulsing and H2O2 treatment compared with untreated RASMC (2.3 ± 0.4 fold-change; P < 0.05; n = 3; Fig. 5, A and E). In RASMC cotreated with CSE and H2O2, rHSP70 at doses of 50 and 100 ng/ml did not significantly affect p21 expression.
Condition media from the treatment groups were collected and used to treat confluent, 48-h serum-starved RASMC. There was an increase in HSP70 expression in cells treated with condition medium from CSE+H2O2 and CSE+H2O2+HSP70 (both 50 and 100 ng/ml doses) compared with serum-free condition medium. Decreased pERK expression was observed in cells treated with condition medium from CSE+H2O2 and CSE+H2O2+HSP70 compared with serum-free condition medium (see Supplemental Fig. 1 in the online version of this article).
Epidemiological studies suggest that cigarette smoke is a strong risk factor for increased carotid intimal thickening or stroke(4, 16, 17, 29).We have previously reported that exposure to cigarette smoke is associated with a significant increase in intimal thickening after arterial wall injury in mice (1, 35). The specific mechanism of the detrimental effects of cigarette smoke on the vascular wall is unknown. Preliminary work with gene microarray analysis in mice exposed to cigarette smoke leading to our study showed a several-fold reduction of HSP70 gene expression. In humans, levels of circulating HSP70 were reported to be inversely related to the progression of intimal/medial thickness in subjects with hypertension (32), identifying it as a potential candidate for intervention.
In this study, we confirmed that exposure to cigarette smoke is associated with reduced arterial wall HSP70 expression in response to injury compared with expression in mice exposed to air. HSP70 is a member of a large family of HSPs that have a protective role in cells under stress. They are mainly intracellular molecules but are released in response to stress (31, 38). HSP70 is one of the more extensively studied HSPs with many functions such as molecular chaperoning and protection against ischemic injury (3, 12, 30). Arterial tissues respond to thermal stress by upregulating the production of HSP70. Thermal treatment of rats enhanced HSP72 expression in the media of injured arteries. This medial HSP expression indicated that HSP72 acted against smooth muscle cell (SMC) activation and migration to the intima and prevented intimal thickening, in part through suppression of oxidative stress (28).
Passive release of HSP70 occurs during cell death. Increasing evidence also suggests active extracellular release of HSP70 by lipid rafts and exosomes (2) from cells exposed to stressors. The specific role is unclear, but reports have shown biological activity of extracellular HSP70. Levels of circulating HSP70 were reported to be inversely related to the progression of intimal/medial thickness in subjects with hypertension (32), suggesting that HSP70 has a protective role in vascular processes. Insufficient HSP70 accumulation in the SMCs localized near necrotic areas of atherosclerotic lesions may lead to plaque rupture (15). In our study, injury to the arterial wall resulted in increased HSP70 expression. Exposure to cigarette smoke in vivo attenuated HSP70 expression and was associated with increased intimal thickening. These findings suggest that attenuation of HSP70 expression is one of the mechanisms by which exposure to cigarette smoke modulates the vascular response to injury. The attenuated expression appeared to be systemic since aortic, as well as splenic HSP70 mRNA expression, was also reduced in mice exposed to cigarette smoke. However, it appears that the decreased HSP70 expression has no obvious physiological effect unless injury of the right carotid artery is performed. Uninjured arteries of the mice exposed to cigarette smoke have no increase in intimal thickening.
Activation of ERK is a key step in the arterial response to injury. ERK activation is followed by an increase in c-jun and c-fos gene expression and enhanced activator protein-1 (AP-1) DNA-binding activity (11, 18). Inhibition of ERK activation attenuates neointimal formation in injured rat arteries (13). ERK activation is involved in the receptor tyrosine kinase pathway signaling. However, reactive oxygen species do not directly activate receptor protein kinases and instead activate cytoplasmic tyrosine kinases to signal mitogenic responses. ERK activation and increased HSP70 expression by H2O2 are partially mediated by the JAK/STAT pathway (26).
In our report, oxidative stress induced by H2O2 in vitro resulted in increased ERK activation and HSP70 expression in RASMC. Pulse treatment of RASMC with CSE prior to H2O2 treatment further increased ERK activation but decreased HSP70 expression. The increased ERK activation was coupled with increased PCNA and p21 expression, and increased percentage of cells in S-phase. The regulatory interaction between pERK and HSP70 has been described previously. pERK is involved in repressing transcriptional activity of heat shock factor-1 on the HSP70 promoter (5). Deficiency of HSP70 in embryonic fibroblasts leads to increased pERK in response to stress (21). Thus, it appears that pERK and HSP70 regulate each others' activation. It is interesting to note that CSE treatment alone did not affect pERK, HSP70, and PCNA expression, suggesting that CSE acts only in synergy with other stressors.
The regulatory pathway described above is through intracellular interaction. Bag-1 is a HSP70 binding protein involved in cell growth signaling in response to stress. Bag-1 binds to Raf-1 kinase activating the downstream ERK pathway promoting cell growth. Overexpressed HSP70 competes for Bag-1, displacing Bag-1 from the Bag-1/Raf-1 complex and arrests DNA synthesis (33). However, it is unclear whether the pERK-HSP70 regulatory loop is similarly affected by the novel role recently proposed for extracellular HSP70.
In our study, at 7 days after cuff injury, rHSP70 treatment of mice exposed to cigarette smoke resulted in decreased pERK expression in injured arteries compared with injured mice exposed to cigarette smoke without treatment. Reduction of pERK expression resulted in reduced intimal thickening 21 days after injury. Bag-1 expression was not different in the two groups of mice. The effect of rHSP70 treatment on intimal thickening was specific in mice exposed to cigarette smoke. Because mice exposed to air did not have reduced pERK expression and intimal thickening after rHSP70 treatment, it suggests a specific inhibitory role of HSP70.
Exogenous rHSP70 directly affects the pERK-HSP70 regulatory loop in RASMC treated with CSE and H2O2 by reducing pERK expression. Consequently, PCNA expression and percentage of cells in S-phase were reduced. The reduction in PCNA expression by rHSP70 treatment was not coupled with significant changes in p21 expression (8). A report by Luo et al. (25) showed similar results with human fibroblast-like synoviocytes (FLS) isolated from patients with rheumatoid arthritis. In their study, FLS stimulated with TNF-α had reduced ERK activation and decreased inflammatory signaling with increasing amounts of HSP70. Thus, the effects of exogenous HSP70 on ERK activation is not limited to cellular stress caused by reactive oxygen species but similarly affects ERK activation through inflammatory signaling. In agreement with our study, treatment with HSP70 alone had no effect. The results of our in vitro experiments with RASMC support the in vivo findings in mice and confirm the findings by Luo et al. (25) in human FLS. How extracellular HSP70 modulates ERK activation remains unknown. Reports have suggested that endogenously expressed HSP70 mediates increased activity of MAP kinase phosphatase-1 (22, 37). It is unclear whether the HSP70 expressed were released into extracellular space and whether the pathway is similar to treatment with exogenous HSP70.
A recent report underscores the complexity of the effects of exposure to cigarette smoke. Treatment of primary lung fibroblasts with water-soluble extract of cigarette smoke for 24 h resulted in increased expression of HSP70 (24). Several factors may cause the difference in our observed results, including the length of time of cell exposure to the extract. In addition, the possibility that there are species differences cannot be excluded. The experiments in our study were initially performed with mouse aortic smooth muscle cells, but the low yield of cells during isolation and the large amount of protein needed to detect HSP70 were limiting factors. We, therefore, switched to RASMC for the in vitro experiments. We did confirm that consistent with the decreased HSP70 expression in RASMC by cigarette smoke extract, mouse aortic smooth muscle cells also had decreased HSP70 expression after pulse exposure to CSE (see Supplemental Fig. 2 in the online version of this article). In addition, our in vitro results with RASMC are consistent with our in vivo findings.
In conclusion, our study suggests that decreased arterial HSP70 expression is an important mechanism by which exposure to cigarette smoke augments intimal thickening. Recombinant HSP70 has a significant beneficial effect in preventing augmented intimal thickening in mice exposed to cigarette smoke through decreased activation of the ERK signal pathway.
Perspectives and Significance
The link between exposure to cigarette smoke and various diseases is well established. What have been critically lacking are studies on the mechanisms of action by this exposure that lead to disease. Dissecting this will inevitably lead to better understanding of the process, as well as better targeted therapies specific to the mechanism of action. We approached the problem by using an unbiased selection process using DNA microarray technique to identify a critical gene that was adversely affected by chronic in vivo exposure to cigarette smoke. We then attempted to reverse this detrimental effect using a therapeutic approach by reconstituting the gene product, HSP70, into the experimental system. We successfully reversed the detrimental effect of cigarette smoke exposure on the vessel wall and performed studies to determine its mechanistic pathway. The effect of exogenously administered rHSP70 is through regulation of ERK activation, and this is specific only to cells or subjects that have been exposed to cigarette smoke or its derivatives.
This study was supported by the Entertainment Industry Foundation and the Heart Fund at Cedars-Sinai Medical Center in Los Angeles, California.
↵* These authors contributed equally to this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society