Gene Transfer of Dominant-Negative Mutants of Extracellular Signal–Regulated Kinase and c-Jun NH2-Terminal Kinase Prevents Neointimal Formation in Balloon-Injured Rat Artery

Balloon injury causes vascular smooth muscle cell (SMC) proliferation and migration and the accumulation of extracellular matrix, which finally result in the neointimal formation.¹² Accumulating evidence indicates that intimal thickening by balloon injury is mediated by a complex interaction of a variety of growth-regulatory molecules, such as growth factors, vasoactive peptides, inflammatory cytokines, or chemokines.³⁴⁵ These molecules are therefore regarded as the useful target for treatment of vascular thickening. However, it has been recently proposed that targeting the critical intracellular signaling steps common to the induction and action of these growth-regulatory molecules may be more useful than the inhibition of a single growth-regulatory pathway, and this notion is supported by recent experimental findings on the prevention of neointimal hyperplasia by in vivo transfer of antisense cdc2 kinase and proliferating cell nuclear antigen oligonucleotides or retinoblastoma protein RB, cyclin kinase inhibitor protein p21, or the dominant-negative mutant of ras gene, as reviewed.⁶⁷⁸ Despite the critical and diverse role of intracellular protein kinase cascades in cellular responses by the above-mentioned neointima-generating molecules, the role of the protein kinases in vascular thickening remains to be determined.

Extracellular signal–regulated kinase (ERK) and c-Jun NH2-terminal kinase (JNK), belonging to the mitogen-activated protein kinase (MAPK) family, are commonly activated by vascular remodeling–related molecules and play the central role in the initiation of cellular responses, including cellular gene expression, growth, migration, or apoptosis.⁹¹⁰¹¹ Furthermore, the possible role of these MAPKs in various cardiovascular diseases has been proposed, as reviewed.¹¹ Recently, we¹² and other groups of investigators¹³¹⁴ have found that both ERK and JNK are rapidly and transiently activated in balloon-injured artery. Despite the in vitro evidence for the involvement of ERK in cultured vascular smooth muscle cell (SMC) proliferation, in vivo administration of PD98059, a selective inhibitor of MAPK/ERK-1 (MEK1), in balloon-injured rat artery failed to prevent intimal SMC proliferation, suggesting no contribution of ERK to intimal hyperplasia.¹⁴ Thus, the role of ERK in the neointima formation in vivo is unclear. Accumulating in vitro evidence indicates that JNK is implicated in various, often opposing cellular responses, as reviewed.¹⁵ For example, there is considerable evidence that JNK activation promotes cell apoptosis, whereas there are also many reports that JNK activation instead inhibits cell apoptosis and promotes cell proliferation or differentiation. Thus, JNK has diverse functions, and its function strongly depends on the cell type and the context of other regulatory influences that the cell is receiving. However, as in the case of ERK, the in vivo role of JNK activation in vascular diseases remains to be elucidated.

In the present study, to elucidate the role of increased ERK and JNK activities after balloon injury, dominant-negative mutants of ERK and JNK were produced and in vivo transfected into rat carotid artery before balloon injury using the gene-transfer method with hemagglutinating virus of Japan (HVJ).^1617181920 We obtained the first evidence that both ERK and JNK are directly involved in neointimal formation after balloon injury.

Materials and Methods

Animals

All procedures were in accordance with institutional guidelines for animal research. Nine- to ten-week-old male Sprague-Dawley rats (Clea Japan, Tokyo, Japan) weighing ≈300 to 350 g were used.

Dominant-Negative Mutant of ERK and JNK

Rat wild-type p44ERK cDNA (W-ERK) and p46JNK cDNA(W-JNK) were produced by polymerase chain reaction (PCR) and confirmed by DNA sequencing. Dominant-negative mutant of p44ERK cDNA (DN-ERK) was produced by PCR using primers designed to produce a lysine (AAG) →arginine (CGG) substitution at lysine 72 in the ATP-binding site of W-ERK.²¹ The mutation was confirmed by DNA sequencing. The dual activating phosphorylation sites in p44ERK (T203 and Y205) were not mutated. Like DN-ERK, a dominant-negative mutant of p46JNK cDNA (DN-JNK) was produced by PCR using primers designed to produce a lysine (AAG) →arginine (CGG) substitution at lysine 52 in the ATP-binding site of W-JNK.²² The dual-activating phosphorylation sites in p46JNK (T183 and Y185) were not mutated. DN-ERK and DN-JNK cannot transfer phosphate and so have negligible catalytic efficiency of the enzymes (see online Figures 1 and 2 available in the data supplement at http://www.circresaha.org), as described.²¹²² W-ERK, W-JNK, DN-ERK, and DN-JNK were tagged hemagglutinin (HA) in the site of NH2 terminus.

W-ERK, W-JNK, DN-ERK, and DN-JNK with HA tag were ligated into the blunted EcoRI site of pUC-CAGGS expression plasmid (pUC/W-ERK, pUC/W-JNK, pUC/DN-ERK, and pUC/DN-JNK, respectively). The pUC-CAGGS vector lacking cDNA served as control vector (vehicle).

Preparation of HVJ Liposomes

The preparation of HVJ liposomes has been described in detail previously.^16171920

In Vivo Gene Transfer and Balloon Injury

The first experiments were performed to examine the effects of DN-ERK and DN-JNK on neointimal formation after balloon injury. Rats were anesthetized with sodium pentobarbital (40 mg/kg IP). The left common carotid artery 20 mm proximal to the carotid bifurcation and the left internal carotid artery at the orifice were temporary occluded by aneurysmal clips. The external carotid artery was ligated at the exposed distal end. To achieve the in vivo gene transfer, the common carotid artery segment was isolated by temporary clips. An infusion cannula was introduced into the segment of the common carotid artery through a small window opened in the external carotid artery. The HVJ-liposome complex with pUC/DN-ERK, pUC/DN-JNK, or pUC-CAGGS vector as control was infused into the closed luminal segment of common carotid artery and incubated for 15 minutes at room temperature, as previously reported.¹⁷¹⁹²⁰ After incubation, the infused fluid of HVJ-liposome complex was removed, the external carotid artery was ligated at the orifice, the blood flow to the common and internal carotid arteries was restored by releasing the clips, and the wound was closed. At 2 days after gene transfer, the endothelial denudation of the left common carotid artery was carried out by 3 passages of a Fogarty 2F balloon catheter (Baxter Healthcare) through the left femoral artery, as previously described.¹²²³ In separate experiments, to examine the synergy of DN-ERK and DN-JNK, we compared the effect of gene transfer of combined DN-ERK and DN-JNK on intimal hyperplasia with that of DN-ERK or DN-JNK alone at 14 days after balloon injury.

The second experiments were performed to examine the effects of W-ERK and W-JNK on neointimal formation after injury. The HVJ-liposome complex with pUC/W-ERK, pUC/W-JNK, or pUC-CAGGS vector (control) was incubated within the lumen of left common carotid artery for 15 minutes in the same manner as the first experiments. At 2 days after gene transfer, the endothelial denudation of the left common carotid artery was carried out, as previously described.¹²²³

Preparation of Arterial Protein Extracts for Protein Kinase Assay

Our methods have been described in detail previously.¹² Briefly, carotid arteries in each group were pooled from 4 to 5 rats to minimize animal-to-animal and procedure variability, and the pooled arteries were homogenized in lysis buffer.¹²

Measurement of Arterial ERK and JNK Activities

ERK and JNK activities were measured by using an in-gel kinase method, as previously described in detail.¹²²⁴²⁵

Western Blot Analysis

Western blot analysis was performed as previously described in detail.¹² Briefly, arterial protein extracts (60 μg protein), prepared as described above, were boiled for 5 minutes in Laemmli sample buffer and then electrophoresed on a SDS polyacrylamide gel (12%), and the separated proteins were electrophoretically transferred to Hybond-PVDF membranes (Amersham Life Sciences). Nonspecific background was blocked by incubating the membrane with 5% bovine serum albumin in Tris-buffered saline (pH 7.6) containing 0.1% Tween 20 (TBS-T) at 4°C overnight. The membrane was then incubated with rabbit polyclonal anti-HA antibody (Covance, Research Products, Inc) (5 μg/mL) for 1 hour at room temperature, washed 4 times with TBS-T, and then incubated with horseradish peroxidase–conjugated donkey anti-rabbit immunoglobulin (Amersham). After additional washing with TBS-T, the membrane was treated with ECL reagent (Amersham), and chemiluminescence was detected by exposure to Hyperfilm-ECL. Blots were quantified using model GS-700 imaging densitometer with Macintosh Software for Image Analysis Systems (Multi-Analyst Version 1.0) (Bio-Rad Laboratories).

Measurement of Intimal/Medial Area Ratio

For the estimation of intimal/medial area (I/M) ratio, rats were anesthetized with ether, carotid arteries were fixed by perfusion of 10% (wt/vol) formaldehyde for 15 minutes under constant pressure, and the middle portion of the left common carotid artery (6-mm length) was removed from each animal, divided into six 1-mm-thick specimens, embedded in paraffin, and sectioned at 3-μm thickness. The 3-μm cross sections were stained with Elastica Van Gieson and H&E. The intimal and the medial areas of each cross section were measured with a light microscope connected to the image analyzer in a blinded fashion. In each animal, the mean intimal and medial areas of each artery were calculated from these 6 sections.

BrdU Immunohistochemistry and Morphometric Analyses

We measured the percentages of BrdU-positive cells at 7 days after injury and the nuclei number at 14 days, as previously reported.²⁶ In brief, for BrdU immunohistochemistry, rats were intraperitoneally injected with 100 mg/kg BrdU at 24 hours and 1 hour before being killed. In the same way as the estimation of I/M ratio, the left carotid arteries were fixed by perfusion of 10% (wt/vol) formaldehyde for 15 minutes, removed, embedded in paraffin, and sectioned at 3-μm thickness. In each animal, 6 sections were examined in a blinded fashion. BrdU immunohistochemistry was performed with a mouse anti-BrdU monoclonal antibody (Amersham) and LSAB2 kit (Dako Japan Co, Ltd).

Total cell number of the intima and media at 14 days after injury was estimated for each section by multiplying the intimal and medial cross-sectional areas, respectively, by the number of nuclei per square millimeter, which was calculated by counting nuclei and is in three representative high-power fields (×200), as previously described.²⁶

Direct Effects of Dominant-Negative Mutant of ERK and JNK on Vascular Smooth Muscle Cell Proliferation In Vitro

Recombinant replication–defective E1- and E3-adenoviral vectors expressing DN-ERK or DN-JNK were constructed using an adenovirus expression vector kit (Takara Biomedicals) according to the method of Miyake et al.²⁷ cDNA encoding DN-ERK or DN-JNK was placed into a cassette cosmid vector PaxCAwt, which possesses CAG promoter comprising a cytomegalovirus enhancer, chicken β-actin promoter, and rabbit β-globin poly A signal. A recombinant adenovirus was constructed by in vitro homologous recombination in 293 cells using the above cosmid vector PaxCAwt containing DN-ERK or DN-JNK cDNA and the adenovirus DNA-terminal protein complex. Recombinant adenoviruses containing bacterial β-galactosidase gene was also constructed as negative control of DN-ERK or DN-JNK in the same way. The titer of the virus was determined by limiting dilution in 293 cells and expressed as plaque-forming units.

Rat aortic SMCs were prepared from thoracic aortas of male Sprague-Dawley rats by using the collagenase digestion method and cultured, as described.²⁸ For all experiments, rat aortic SMCs from passages 4 to 7 were used. SMCs were grown to 70% to 80% confluence and then made quiescent by incubation with DMEM containing 0.1% FBS for 48 hours and stimulated with addition of 2% serum. In vitro gene transfer to aortic SMCs was carried out by incubation with the adenoviral vector with multiplicity of infection of 50 in DMEM containing 0.1% FBS for 1 hour at 37°C, 5% CO₂, and 95% air. Then SMCs were made quiescent for 48 hours before being assessed for the expression and the effect of the transferred gene. SMCs in 6-well plates were stimulated by 2% serum for 19 hours and pulsed with 1 μCi/mL [³H]thymidine for 5 hours. Then cells were washed twice with PBS, incubated for 5 minutes in 5% trichloroacetic acid, washed by methanol, and dissolved in 99% formic acid. The incorporation of [³H]thymidine into trichloroacetic acid–insoluble material was measured by liquid scintillation spectrophotometer. For the assay of cell growth, SMCs in 60-mm plates were stimulated by 5% serum, and cell number was counted by Coulter counter (Beckman).

Statistical Analysis

All data are presented as mean±SEM. Statistical significance was determined with one-way ANOVA followed by Duncan multiple-range comparison test using Super ANOVA (Abacus Concepts, Inc). Differences were considered statistically significant at a value of P<0.05.

Results

Effects of Dominant-Negative Mutants on Arterial ERK and JNK Activities In Vivo

To distinguish between the dominant-negative mutant and endogenous kinase, we used DN-ERK–encoded and DN-JNK–encoded HA-tag in the site of its NH2 terminus. To confirm the successful gene transfer of these dominant-negative mutants into carotid artery in vivo with HVJ-liposome method, carotid arteries, subjected to gene transfer of pUC/DN-ERK, pUC/DN-JNK, or pUC-CAGGS before 2 days, were removed at 5 minutes and 2, 4, and 7 days after balloon injury. As shown in Figure 1A, the Western blot analysis with anti-HA antibody showed the single band corresponding to the same molecular weight as p44ERK (44 kDa) in arterial protein extracts subjected to gene transfer of DN-ERK but no band in those of pUC-CAGGS (control vector), confirming the successful expression of DN-ERK within arterial cells at 2, 4, and 7 days after injury. Being in good agreement with our previous report,¹² p44ERK and p42ERK were remarkably activated at 5 minutes after injury, and this activation was significantly inhibited by gene transfer of DN-ERK (P<0.01).

As in the case of DN-ERK gene transfer, Figure 1B showed that DN-JNK gene transfer allowed for the significant expression of the protein in arterial tissues at 2, 4, and 7 days after injury, and the peaked activation of arterial JNK at 5 minutes after injury was successfully suppressed by gene transfer of DN-JNK(P<0.01).

Effects of Dominant-Negative Mutants of ERK and JNK on Neointimal Formation

As shown in Figures 2 and 3, remarkable neointimal formation was observed in the vehicle-transfected group at 14 and 28 days after injury. DN-ERK and DN-JNK significantly decreased the I/M ratio (0.62±0.09 and 0.43±0.09, respectively) compared with vehicle (pUC-CAGGS) (0.97±0.05) (P<0.01; Figure 3A) at 14 days. Even at 28 days, DN-ERK and DN-JNK significantly suppressed neointimal formation (P<0.01; Figure 3B). On the other hand, DN-ERK or DN-JNK did not affect the medial areas at 14 or 28 days.

Total intimal nuclei number per section in arterial segments transfected with DN-ERK (891±77) or DN-JNK (857±90) was similar and significantly smaller than that of vehicle-transfected arterial segment (1808±134) (P<0.01). There was no significant difference among the three groups transfected with vehicle, DN-ERK, and DN-JNK with respect to medial nuclei number (681±28, 711±36, and 735±37, respectively).

In separate experiments, we compared the effect of combined DN-ERK and DN-JNK gene transfer on intimal hyperplasia with that of each mutant alone at 14 days after balloon injury. The I/M ratio at 14 days after balloon injury was 0.95±0.06, 0.69±0.05, 0.64±0.06, and 0.52±0.05 in vehicle and groups treated with DN-ERK, DN-JNK, and combined DN-ERK and DN-JNK, respectively (each n=7). DN-ERK, DN-JNK, and their combination all significantly reduced the I/M ratio compared with vehicle (P<0.01). Although the combination of these genes decreased the I/M ratio more than DN-ERK alone (P<0.01), there was no significant difference between DN-JNK alone and combined DN-ERK and DN-JNK.

Effects of DN-ERK and DN-JNK on SMC Proliferation In Vivo

To estimate SMC proliferation, we measured the intimal and medial BrdU index at 7 days after injury. As shown in Figures 4 and 5A, the percentages of BrdU-positive cells in the neointima from groups transfected with DN-ERK (20.2±2.2%) or DN-JNK (18.7±3.2%) were smaller than those of the control group (36.9±3.9%) (P<0.01).

As shown in Figure 5B, there was no BrdU-positive cell in the media of noninjured arteries. The percentages of BrdU-positive cells in the media from groups transfected with DN-ERK (3.3±0.4%) and DN-JNK (3.2±0.6%) were smaller than the percentage of the control group (8.6±1.2%) (P<0.01).

Effects of Wild-Type ERK and JNK on Neointimal Formation

We examined the effects of W-ERK or W-JNK transfection on neointimal formation after balloon injury. As shown in Figures 6 and 7A, W-ERK and W-JNK transfection (1.19±0.09 and 1.21±0.11, respectively) significantly increased the I/M ratio compared with pUC-CAGGS transfection (0.76±0.06). Medial areas were not affected by transfection of W-ERK, W-JNK, or both (Figure 7B).

Inhibitory Effects of Dominant-Negative Mutant ERK and JNK on Vascular Smooth Muscle Cell Proliferation In Vitro

To confirm that ERK and JNK are directly involved in vascular SMC proliferation, we examined the effects of DN-ERK and DN-JNK on vascular SMC proliferation in vitro. Figure 8 shows the effect of these dominant-negative mutants on the kinase activity, ³H-thymidine incorporation, and cell number in cultured rat aortic SMCs stimulated with serum. Stimulation of aortic SMCs with 2% serum significantly activated ERK and JNK by 6.4-fold (n=3, P<0.01) and 9.9-fold (n=3, P<0.01), respectively, and increased both ³H-thymidine incorporation and cell numbers (P<0.01). These increases in ERK and JNK activities were inhibited by DN-ERK by 90% (n=3, P<0.01) and DN-JNK by 67% (n=3, P<0.01), respectively (Figures 8A and 8B, respectively). DN-ERK and DN-JNK inhibited serum-induced increase in [³H]thymidine incorporation by 51% and 60%, respectively (Figure 8C) and also attenuated the increase in cell number by 86% and 89%, respectively (Figure 8D). On the other hand, LacZ gene transfer did not affect ERK or JNK activity, the increase in [³H]thymidine incorporation, or cell number in aortic SMCs stimulated with serum.

Discussion

The major findings of this work were that gene transfer of either DN-ERK or DN-JNK significantly prevented neointimal formation after balloon injury and inhibited vascular SMC proliferation in vitro and in vivo, indicating that both ERK and JNK are involved in intimal hyperplasia in vivo.

We and other groups of investigators have reported that vascular ERK and JNK are significantly activated by balloon injury,¹²¹³¹⁴ the injection of angiotensin II²⁴ or other hypertensive agents,²⁹ and chronic hypertension.³⁰ All these findings led us to propose that the activation of ERK and JNK may trigger a series of complex molecular events leading to the neointimal formation. However, the investigation of the in vivo blockade of these MAPKs has been hampered by the absence of their specific and potent pharmacologic inhibitors, and the significance of activation of these kinases in vascular diseases in vivo remains to be defined.

In the present study, to specifically inhibit the activation of vascular endogenous ERK and JNK in vivo, we produced dominant-negative mutants of ERK and JNK. To transfect these mutant genes into rat carotid artery, we used the HVJ-liposome technique,^16171920 because the use of HVJ-liposome method allows for the effective gene transfer into medial SMCs of the intact carotid artery not subjected to endothelial denudation.³¹ To distinguish the transfected dominant-negative mutant of ERK or JNK from the endogenous kinase, the dominant-negative mutants were HA-tagged. Western blot analysis with anti-HA antibody and in-gel kinase assay confirmed that DN-ERK and DN-JNK were effectively expressed within the arterial cells and the expressed mutants could potently inhibit the activation of their respective kinase after balloon injury. Thus, our present method allowed us to elucidate the direct role of these MAPKs in neointimal formation after balloon injury.

Despite in vitro solid evidence for the involvement of ERK in cell growth, the role of ERK in vivo is poorly understood. In the present study, we obtained the first evidence that the inhibition of ERK activation by gene transfer of DN-ERK significantly suppressed the neointimal formation, whereas wild-type ERK gene transfer accelerated the neointima thickening, confirming the contribution of ERK activation to neointimal formation after balloon injury. As shown by labeling index of BrdU-positive nuclei and total cell counting, gene transfer of DN-ERK suppressed SMC proliferation in either the intima or the media and reduced intimal cell number. This observation, taken together with the in vitro results that DN-ERK inhibited cultured vascular SMC proliferation (Figure 8), demonstrated that the suppression of SMC proliferation was involved in the inhibition of intimal thickening by DN-ERK. However, additional study is needed to elucidate the precise mechanism underlying the inhibition of neointima, because the possible contribution of migration or apoptosis cannot be excluded.

On the other hand, Koyama et al,¹⁴ who examined the effects of PD98059, a selective inhibitor of MEK-1 on SMC proliferation after balloon injury of rat carotid artery, reported that PD98059 blocks medial cell replication but not intimal cell replication in injured artery, being in disagreement with our present data on significant inhibition of replication of intimal cells as well as medial cells by DN-ERK. However, because this group of investigators did not examine the inhibitory effect of PD98059 on arterial ERK activity at any earlier time points than 30 minutes after balloon injury, it was uncertain in their study whether PD98059 could inhibit arterial ERK activation at the peak time point (5 minutes) after balloon injury. Furthermore, because PD98059 is not such a potent inhibitor, it is unclear whether this compound is sufficiently effective in in vivo conditions.³² Thus, the failure of inhibition of SMC proliferation in the intima after balloon injury by PD98059 is possibly attributable to the insufficient blockade of ERK activation by this compound in vivo. Moreover, it cannot be excluded that MEK inhibition with PD98059 may not only inhibit ERK activation but also affect other protein kinase cascades.³³ Thus, our present study provided the first evidence for the important role of ERK activation in neointimal hyperplasia after balloon injury.

JNK, belonging to another MAPK family, activates c-Jun by phosphorylation at its Ser 63/73³⁴ and has different upstream and downstream cascades from ERK.⁹¹⁰¹¹ As recently reviewed,¹⁵ JNK activation has diverse function. For example, the activation of the JNK pathway has been reported to play a positive role in cell growth in fibroblasts, T cells, and cardiac myocytes, whereas other studies have shown the negative role of JNK in cell proliferation and the apoptotic function of JNK in hippocampal neurons or PC12 cells. Therefore, whether JNK activation leads to cell proliferation or apoptosis depends on the cell type and the signaling pathways that are simultaneously activated.¹⁵ However, the significance of JNK activation in injured artery in vivo remains unknown. In the present study, of note are the observations that gene transfer of DN-JNK prevented SMC proliferation and neointimal formation after balloon injury, whereas wild-type JNK gene transfer enhanced neointima formation. Furthermore, our present work showed that DN-JNK suppressed cultured vascular SMC proliferation in vitro. These results demonstrated that as in the case of ERK, JNK activation participated in neointimal formation, at least in part by stimulating SMC proliferation. As in DN-ERK, DN-JNK gene transfer did not alter the medial area despite the suppression of medial SMC proliferation. These findings may be attributable to the fact that the medial area is affected not only by medial SMC proliferation but also by other events, such as medial SMC apoptosis, extracellular matrix accumulation, and SMC migration from the media to the intima, although additional study is needed to elucidate this point.

Study Limitation

As in the case of most previous studies concerning the effects of in vivo gene transfer with adenovirus vector or HVJ liposome on arterial hyperplasia,⁶ in this work, the duration of isolation of arterial segment for gene transfer was longer than that accepted in clinical situation. Therefore, the development of more effective gene transfer technique is essential for the clinical application. Furthermore, arterial restenosis in humans after balloon injury generally occurs at longer periods than in rats. Thus, additional study is needed to determine whether our present findings can apply to human restenosis.

In conclusion, we obtained the first evidence that specific blockade of ERK or JNK activation by gene transfer of their dominant-negative mutants prevents SMC proliferation and neointimal thickening in balloon-injured artery, demonstrating that either ERK or JNK activation triggers a series of molecular events leading to neointimal hyperplasia in vivo. Thus, our work provides new insight into the molecular mechanism of neointimal hyperplasia. We propose that these MAPKs are the new therapeutic targets for the prevention of vascular diseases. However, additional study is needed to elucidate the detailed mechanism underlying neointimal hyperplasia induced by these MAPKs.

Original received November 27, 2000; revision received April 10, 2001; accepted April 10, 2001.

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Footnotes