The contribution of atypical protein kinase C (PKC)-ζ to ANG II-accelerated restenosis after endoluminal vascular injury was investigated by using the rat carotid balloon injury model. Exposure of injured arteries to ANG II resulted in an extensive neointimal thickening (1.9 times) compared with vehicle at day 14. Treatment with PKC-ζ antisense, but not scrambled, oligonucleotides reduced neointimal formation observed in the presence or absence of ANG II. Examination of early events (2 days) after injury showed an increase in cellularity in the perivascular area of the artery wall that was transferred to the adventitia and media after exposure to ANG II, events blocked by PKC-ζ antisense, but not scrambled, oligonucleotides. A positive correlation between medial cellularity at day 2 and extent of neointimal growth at day 14 was established. Immunohistochemical analysis showed that upregulation of inflammatory markers after injury, as well as infiltration of ED1+ monocytes/macrophages from the perivascular area to the adventitia, was accelerated by ANG II. However, ANG II-stimulated medial increase in cellularity was proliferation independent, and these cells were monocyte chemoattractant protein-1+/vimentin+ but ED1−/VCAM−. PKC-ζ is degraded after injury, and inhibition of its neosynthesis in medial vascular smooth muscle cells or in infiltrating cells with PKC-ζ antisense attenuated medial cellularity and expression of inflammation mediators without reversing smooth muscle cell dedifferentiation. Together, these data indicate that PKC-ζ plays a critical role in normal and ANG II-accelerated neointimal growth through a mechanism involving upregulation of inflammatory mediators, leading to cell infiltration in the media of the vascular wall.
- protein kinase C-ζ
- smooth muscle cells
changes in vascular wall structure underlie diseases such as atherosclerosis, vasculitis, systemic and pulmonary hypertension, and accelerated arteriosclerosis (58). Vascular wall remodeling seen during restenosis in response to injury involves vascular smooth muscle cells (VSMC) and adventitial fibroblast proliferation and migration (11, 26). Vascular lesions are also associated with intimal hyperplasia (14). Recent studies (28, 55, 58) have characterized atherosclerosis as an inflammatory disease because early adhesion, infiltration, and migration of leukocytes and increase in leukocyte-selective adhesion molecules and proinflammatory cytokines in the arterial wall precede and are required for neointimal formation.
ANG II, a vasoactive agent, present in the circulation and also synthesized locally, promotes growth of intimal endothelial cells, medial VSMC, and adventitial fibroblasts (50). ANG II causes hypertrophy of VSMC and cardiac myocytes and hyperplasia of cardiac and adventitial fibroblasts through stimulation of the ANG II type 1 receptor subtype (21, 43, 44). ANG II promotes a marked increase in the number of macrophages present in the adventitial tissue underlying lesions and potentiates adventitial fibrosis (6, 8, 9). Moreover, the effects of ANG II are augmented after vascular injury (8, 22, 42), and atherosclerosis is accelerated in ANG II-infused rodent models (9). ANG II also displays a broad array of proinflammatory effects that may directly affect the atherosclerotic process, such as prothrombic activity on platelets (12), leukocyte recruitment, activation of macrophages (42, 50), and direct binding of monocytes to the endothelium and VSMC (4, 6). However, the exact sequence of events after vascular injury, especially regarding the role of ANG II on various cell types involved in repair and inflammatory response, is not known.
Protein kinase C (PKC)-ζ, a calcium- and diacylglycerol-independent atypical PKC, is activated by and mediates some of the functional effects of ANG II in VSMC (27, 35). PKC-ζ also mediates the transcription of adhesion molecules in endothelial cells (41), adhesion and chemotaxis of polymorphonuclear leukocytes (24), and their binding to the endothelium (20). The mechanisms responsible for the vessel retention of leukocytes and their role in the initiation of the restenosis are not known. Increased PKC-ζ activity has been linked to atherosusceptibility in regions of aorta normally subjected to flow turbulence (29). Whether PKC-ζ contributes to the proinflammatory and progrowth effects of ANG II that accelerate atherosclerosis and restenosis is not known. The proinflammatory effect of ANG II and other growth factors are mediated by the transcription factor NF-κB (53). NF-κB is required for the upregulation of proinflammatory cytokines, adhesion molecules, and chemokines (32). Interestingly, PKC-ζ regulates NF-κB activation in many cell types, including endothelial cells and leukocytes (13, 30, 41). These observations raise the possibility that the effects of ANG II on vascular growth and remodeling of arteries in vivo are mediated by PKC-ζ. In the present study, we investigated the contribution of PKC-ζ to normal and ANG II-accelerated neointimal growth and the underlying mechanism of its action by using a selective PKC-ζ antisense (AS) oligodeoxyribonucleotide (ODN) in the rat model of carotid balloon injury. In addition, we examined the effect of ANG II and PKC-ζ AS on proliferation, differentiation, and inflammation in the arterial wall.
MATERIALS AND METHODS
Male Sprague-Dawley rats were purchased from Charles River (Wilmington, MA). All protocols were approved by the Institutional Animal Care and Use Committee at the University of Tennessee Health Science Center and were consistent with the Guide for the Care and Use of Laboratory Animals, published by the National Institutes of Health (NIH Publication No. 85-23, Revised 1996).
Carotid arteries or cells were transfected with an 18-mer PKC-ζ AS, GTCGGTCCTGCTGGGCAT, or scrambled (SCR), GGCCGTCGTTCGGTGTAC, oligonucleotide (IDT, Coralville, IA) containing three phosphorothioate bonds at the 5′ and 3′ ends. PKC-ζ AS ODN was designed from the first six codons of rat PKC-ζ cDNA sequence (31) following the recommendations of Stein (49) and Rabbani and Wang (40) for minimizing nonspecific effects in vivo. The ODNs do not share homology with any other known cDNA sequence. The ODNs were complexed for 20 min with Oligofectamine reagent according to the manufacturer's instructions (Invitrogen, San Diego, CA). Transfection efficiency after 48 h was measured in cells and arteries by Western blot analysis using PKC-ζ antibody as previously described (35).
Balloon injury and morphometric analysis.
Balloon injury was performed on the carotid artery of male Sprague-Dawley rats anesthetized with ketamine-xylazine as previously described (60). The left common carotid artery was exposed through a midline cervical incision, and blood flow to the site was temporarily interrupted by ligation of the left common, internal, and external carotid arteries with vessel clips. The uninjured right carotid artery was used as a control. A Fogarty 2-Fr embolectomy catheter (Baxter, Deerfield, IL) was introduced in the common carotid artery through the external carotid branch and passed three times through the clamped section with balloon inflated to denude the endothelium by intraluminal passage. A catheter was introduced through the external carotid artery branch for administration of different agents. The lumen of the injured artery was infused with saline containing Oligofectamine (vehicle) with or without PKC-ζ AS or SCR ODN (200 nM) and in the presence or absence of ANG II (100 nM) for 1 h in a final volume of 0.5 ml. These different solutions were prepared as described in Transfection. Intraluminal incubation for 1 h or a shorter period of exposure to other phosphorothioate AS ODN has been shown to be effective in reducing neointimal growth in the rat carotid injury model (1, 33, 40, 57). After 1 h, the solution was then withdrawn and the external carotid artery was ligated. The blood flow was restored by removing the clips; the incision was sutured, and the animals were given buprenorphine (0.5 mg/kg sc; Greenpark Pharmacy) for analgesia and allowed to recover. Higher concentration (>1 μM) or longer exposure (>1 h) to ANG II resulted in thrombosis, seen after 1 day, or complete occlusion of the blood vessel, seen after 2 wk. Flushing the artery with saline after incubation reduced this occurrence. Thrombi were not observed in the absence of ANG II or in the presence of PKC-ζ AS ODN. Therefore, deendothelialization did not promote thrombosis by itself as reported by Hayashi et al. (19). However, deendothelialization in the presence of ANG II, a thrombogenic agent (12), increased the probability of thrombus formation. We excluded any nonselective effect of Oligofectamine or oligonucleotides on neointimal formation because 1) Oligofectamine was added in all experimental treatments including vehicle and 2) PKC-ζ AS, but not SCR ODN, reduced neointimal formation. After the indicated time (2 or 14 days), animals were euthanized and arteries were perfused through the left ventricle of the heart with saline and then fixed with 10% formalin; the treated segment of carotid artery, including perivascular connective tissue, was removed. Arteries were dehydrated, embedded in paraffin, and sectioned with a microtome (5 μm). Sections were placed on a glass slide, deparaffinized, and stained with hematoxylin and eosin. A minimum of eight different rats was used per treatment. The surface areas of the intima [lumen to internal elastic lamina (IEL)] and media (IEL to external elastic lamina) were calculated using an Image analysis system (Knontron IBAS 2.0), and the intima-to-media (I/M) ratio of randomly selected sections of carotid arteries was analyzed in a double-blind manner.
All tissue sections on slides were treated with antigen unmasking solution (Vector, Burlingame, CA) before immunohistochemistry. Tissue sections were processed for immunohistochemical staining with antibodies directed against rat PKC-ζ (Santa-Cruz Biotechnology), rat ED1 that detects monocytes/macrophages/dendritic cells (Research Diagnostics, Flanders, NJ), monocyte chemoattractant protein-1 (MCP-1; RDI), ICAM-1, VCAM-1, vimentin (Santa Cruz), SM α-actin (Sigma), using the Vectastain Elite ABC kit (Vector) containing an anti-rat biotinylated secondary antibody and processed according to the manufacturer's specification. Dilution of antibodies used in our experiments was calculated according to the recommended protocol for immunohistochemistry of each manufacturer. Controls with IgG alone matched for each secondary antibody used were performed to ensure selectivity and remove false positive. Positive expression of several markers (vimentin, ED1, and VCAM) was performed with sections from rat spleen. Color was developed by using 3,3′-diaminobenzidine, which generated a brown reaction product. Slides were counterstained with hematoxylin. Cell proliferation was measured with a bromodeoxyuridine (BrdU) in situ detection kit (BD Pharmingen). BrdU was injected intraperitoneally (10 mg/rat) at 24 and 12 h before euthanasia. Collagen was detected with Picrosirius red stain.
VSMC from male Sprague-Dawley rats were isolated and maintained in medium 199 containing penicillin, streptomycin, and 10% fetal bovine serum under 5% CO2.
Western blot analysis of carotid artery and VSMC.
Carotid arteries were excised 2 days after injury, washed in PBS, stripped of perivascular tissue, and then quickly snap frozen in liquid N2 and ground. The tissue powder was lysed in 200 μl radioimmunoprecipitation assay buffer containing 0.1% Triton X-100 and protease and phosphatase inhibitors, and protein concentration was determined by the Bradford assay. Lysates from VSMC were prepared as previously described (36). Equal amounts of protein were separated by SDS-PAGE and transferred onto nitrocellulose membranes. Blots were incubated with antibodies to PKC-ζ, VCAM, SM22α (Novus), SM α-actin, and β-actin (Sigma). Bands were visualized with the ECL Plus detection system (Amersham Pharmacia, Piscataway, NJ). It should be noted that PKC-ζ antibody also recognizes atypical PKC-ι, a protein that shares a similar molecular weight and is not distinguishable by Western blot analysis (47). PKC-ι is not targeted by PKC-ζ AS, and therefore PKC-ι levels in VSMC do not decrease after treatment with PKC-ζ AS.
Detection of chromatin in tissue sections.
The morphological changes of nuclear chromatin were examined by staining the chromatin with Hoechst 33342 dye. Tissue sections were washed in PBS for 2 h and stained with Hoechst 33342 (5 μg/ml in PBS) for 30 min at 37°C and viewed under fluorescence microscopy (Olympus IX-50 microscope, WU filter). In addition, some sections were stained with hematoxylin alone.
The protein level of PKC-ζ in VSMC was analyzed by densitometric analysis and expressed as a percentage of the level of vehicle-treated cells. Quantitation of PKC-ζ staining (at day 14) was performed with Image J by measuring pixel numbers in six randomly chosen areas (0.01 square-inch surface) in each arterial layer. The nontissue background was subtracted for each measurement. Data for morphometric and densitometric analysis were analyzed by one-way ANOVA and Dunnett's multiple comparisons test. The null hypothesis was rejected for P < 0.05. Scatchard plot analysis was performed for data from medial cellularity at day 2 and I/M ratio at day 14.
PKC-ζ AS inhibits ANG II-induced neointimal formation.
ANG II has systemic (endocrine) and local (paracrine and autocrine) effects, favoring cell growth and differentiation. We studied the local effect of ANG II in injured arteries. Balloon-injured left carotid arteries in rats showed extensive neointimal thickening 2 wk after initial and transient exposure to ANG II (1 h, 100 nM). Treatment with PKC-ζ AS, but not SCR ODN, decreased the neointimal thickening after the postangioplasty remodeling process both in the absence or presence of ANG II (Fig. 1A). Neointimal growth was quantified and expressed as the I/M ratio (Fig. 1B). The I/M ratio increased by 1.9 times from 0.771 with vehicle to 1.471 with ANG II treatment. Exposure of arteries to PKC-ζ AS, but not SCR ODN, suppressed both injury-induced and ANG II-stimulated postinjury neointimal growth. These data indicate that PKC-ζ AS was able to reduce neointimal thickening in balloon-injured carotid artery in the presence or absence of ANG II. The effect of local and transient exposure to ANG II on neointimal growth at day 14 was apparent at lower concentrations of ANG II. The I/M ratio was 1.398 ± 0.139 at 1 nM, 1.613 ± 0.151 at 10 nM, and 1.496 ± 0.272 at 100 nM ANG II. Acceleration of neointimal growth with transient exposure to ANG II was also apparent at day 7 (not shown). We used fluorescent 6-FAM-labeled-PKC-ζ AS (IDT) to visualize transfection in carotid artery and in subconfluent first passage rat VSMC with the use of an Olympus IX50 microscope (Fig. 2). Pictures were taken 24 h after carotid injury or VSMC transfection. Background fluorescence in elastic fibers is visible in the balloon-injured artery as previously reported (23). Transfection with 6-FAM-PKC-ζ AS resulted in widespread fluorescence in the injured medial layer (Fig. 2A). In VSMC, Oligofectamine-containing vehicle did not show any background fluorescence, whereas 6-FAM-PKC-ζ AS transfection showed different intensity of fluorescence in cultured cells (Fig. 2B). The efficiency of PKC-ζ AS was also measured by Western blot analysis on samples from rat VSMC in culture. PKC-ζ AS caused a marked decrease in PKC-ζ levels in a concentration-dependent manner, with maximal inhibition seen at 200 nM. The SCR ODN did not alter PKC-ζ level (Fig. 2D).
Localization of PKC-ζ in arterial wall at day 14 after injury.
Immunohistochemistry analysis performed on injured artery sections without ANG II treatment at day 14 showed higher levels of PKC-ζ in the luminal border and adventitia than in the neointima, whereas medial levels were negligible (Fig. 3A). This pattern of distribution was not altered by PKC-ζ AS or SCR ODN. ANG II produced an increase in PKC-ζ staining in the periluminal area, adventitia, and neointima, in an order similar to that observed with vehicle alone. The intensity of PKC-ζ staining was quantified, and the order of staining in all groups was always luminal border area > adventitia > neointima; low staining was observed in the media (Fig. 3, C and D). PKC-ζ AS treatment reduced PKC-ζ staining levels, compared with vehicle or SCR ODN, in the periluminal and adventitial area of the arterial layers, either in the absence (Fig. 3C) or presence (Fig. 3D) of ANG II. It is important to note that PKC-ζ is present in the media of uninjured arteries (Fig. 3A). Because the medial PKC-ζ level was low and not altered by PKC-ζ AS at day 14 after injury, we hypothesized that the inhibitory effect of PKC-ζ AS on neointimal growth was occurring early in the repair process after vascular injury.
Cellularity and quantitative analysis of media at day 2.
The arterial wall was analyzed for several parameters 2 days after injury, sufficient time for the antisense to deplete PKC-ζ and before any neointimal growth in the vessel (Fig. 4). A fundamental difference was observed between vehicle (injury control) and ANG II-treated rats. Hematoxylin-labeled nuclei were present in higher numbers in the medial area of ANG II- and PKC-ζ SCR ODN-treated carotid arteries (Fig. 4A). In contrast, in vehicle or PKC-ζ AS ODN-treated arteries, there was no change in nuclei numbers in the medial and sub-IEL area. Additional analysis was performed with chromatin labeling by using the fluorescent Hoechst 33342 dye. Fluorescent-labeled nuclei show the same distribution/cellularity as with hematoxylin, with ANG II promoting an increase in cellularity from the perivascular tissue to the adventitial and medial area (Fig. 4A). Higher magnification (×100) of Giemsa-stained sections clearly shows an increased perivascular cellularity after injury (vehicle) and an increased medial and adventitial cellularity after exposure to ANG II (Fig. 4B). High cellularity may indicate an increase in cell number through proliferation or cell migration. Medial cellularity was increased at day 2 by ANG II as shown in Fig. 4. Because ANG II promoted an acceleration of neointimal growth at day 7 (not shown) and day 14, we studied the correlation between I/M ratio at day 14 and medial cell number at day 2. The data of four different treatments, vehicle, ANG II, PKC-ζ AS + ANG II, and PKC-ζ SCR + ANG II, repeated 8 times were plotted in Fig. 5B. Analysis of these data shows that the extent of neointimal growth at day 14 measured by I/M ratio correlates with the medial cellularity measured at day 2 (R2 = 0.60). Therefore, the medial cell number measured early after angioplasty could be used as an indicator of the extent of future neointimal growth.
Effect of ANG II and PKC-ζ on proliferation, differentiation, and inflammation in arterial layers and perivascular tissues.
PKC-ζ, shown in brown color, was detected mainly in the sub-IEL medial layer and in the perivascular tissue of injured arteries treated with vehicle. PKC-ζ was present in the adventitia and media of ANG II- and SCR ODN-treated arteries, whereas AS ODN treatment was negative for PKC-ζ staining in the periluminal area. Thus ANG II caused an increase in PKC-ζ staining and cell number in the adventitia that was reversed by PKC-ζ AS ODN.
Proliferation and migration of adventitial and medial cells are together with inflammation the hallmark of the restenotic response to vascular injury (26, 58). BrdU incorporation at day 2 after injury, shown in brown color, was used as marker for cell proliferation (Fig. 6). BrdU labeling was strong in the perivascular area of injured vessels and in the adventitia of ANG II-treated carotid arteries. However, proliferation of cells in the medial layer that was increased in injured vessels was not altered with ANG II, indicating that increased medial cellularity elicited by acute exposure to ANG II probably involves increased cell migration rather than cell proliferation. In addition, perivascular BrdU labeling was stronger in vehicle and PKC-ζ AS sections than in ANG II-treated arteries, where it was mostly adventitial in location. Modulation of PKC-ζ level with AS ODN decreased adventitial, but not medial, cell proliferation, independent of the inhibition of ANG II-mediated increase in medial cellularity (Fig. 6). Cell proliferation was not detected in uninjured arteries.
Expression of VSMC-selective SM α-actin is decreased after vascular injury (56). Media staining for SM α-actin was decreased after injury alone (Fig. 6). Acute exposure to ANG II or modulation of PKC-ζ expression did not reverse the loss of medial SM α-actin. Vimentin is the main intermediate filament protein in mesenchymal tissues. Expression of vimentin is upregulated during development and in pathological conditions (16). Vimentin staining was increased in injured sections in the media and perivascular area. Exposure to ANG II augmented medial vimentin expression in the media and the adventitia (Fig. 6). It should be noted that the ANG II-induced increase in medial cellularity was positive for vimentin expression. Increased medial vimentin staining could be due to dedifferentiation of local VSMC or migration of other cell types from the perivascular tissue or the luminal side.
Blood cells such as monocytes/macrophages, granulocytes, lymphocytes, and platelets are involved in the initiation and progression of atherosclerosis and restenosis after vascular injury (18, 28, 58). Monocyte and macrophage infiltration is an early event after endoluminal injury (14, 18, 28, 58). Moreover, chronic infusion of ANG II has been shown to promote infiltration of mononuclear inflammatory cells (6, 8, 9). The contribution of PKC-ζ and the effects of ANG II on ED1+ monocytes/macrophages was therefore examined 2 days after injury. Vascular injury promoted an increase in ED1+ cells in the perivascular area of the injured vessels, whereas acute application of ANG II shifted ED1+ macrophages to the adventitial layer, an infiltration that was not reduced by PKC-ζ AS or SCR ODN (Fig. 6). These data indicate that transient exposure to ANG II promoted an acceleration in macrophage infiltration of the vascular wall. ANG II-increased medial cellularity was not ED1+. Attachment and infiltration of leukocytes in the arterial wall is mediated through the expression of chemokines, such as MCP-1, and adhesion molecules, such as ICAM and VCAM (28, 34, 55, 58). Expression of MCP-1 was upregulated in the medial area at day 2 after injury, potentiated with ANG II in both perivascular area and in the vessel wall, whereas PKC-ζ AS reduced its overall expression. ANG II-induced increase in medial cellularity was MCP-1+. VCAM was expressed in the perivascular area of injured vessels and in the adventitia of ANG II-treated vessels in a PKC-ζ-dependent manner (Fig. 6). Medial cells increased by ANG II treatment were negative for VCAM, ED1, and PCNA. ICAM expression was weak, in contrast to MCP-1 or VCAM, and not subjected to changes with regard to ANG II treatment or PKC-ζ modulation (data not shown). Controls with IgG alone matched for each secondary antibody used, and positive expression of markers in the spleen are shown below carotid sections.
Effect of ANG II and contribution of PKC-ζ to smooth muscle differentiation and inflammation in carotid artery.
VSMC undergo phenotypic modulation/switching after vascular injury, characterized by decreased expression of VSMC-selective marker genes such as SM α-actin and SM22, among other genes (56). On the other hand, these VSMC markers are upregulated during myofibroblastic differentiation occurring after vascular injury (5). The expression of these smooth muscle markers in the segment of injured carotid artery without perivascular tissues was measured by Western blot analysis 2 days after injury (Fig. 7A). SM22, a marker of differentiated VSMC, disappeared after angioplasty. The expression of SM α-actin was also reduced after endoluminal injury. Surprisingly, PKC-ζ AS treatment reduced SM α-actin level more than ANG II alone or PKC-ζ SCR. The pattern of PKC-ζ expression determined by Western blot analysis in different groups showed decreased PKC-ζ expression after injury that was reversed with ANG II treatment but inhibited with PKC-ζ AS. PKC-ζ degradation is a common mechanism used for downregulation after its activation by phosphorylation (2, 25, 48). We have found several bands below 75 kDa in our Western blots of injured arteries from vehicle samples and ANG II that are recognized by PKC-ζ antibody, indicating degradation of the protein after apoptosis (46) or enzymatic activation by phosphorylation.
Inflammation was measured by the expression of ED1 and VCAM. ED1 was not detected in uninjured arteries. Vascular injury upregulated ED1 levels, and ANG II further potentiated its expression, independent of PKC-ζ. VCAM expression was upregulated after angioplasty, potentiated with ANG II, and decreased by PKC-ζ AS (Fig. 7B). Expression of β-actin is upregulated in VSMC undergoing dedifferentiation toward a synthetic state (5). In addition, β-actin is also expressed by several cell types, such as leukocytes and fibroblasts, that are weakly represented in the uninjured arterial wall; expression of β-actin is greatly increased after vascular injury. β-Actin expression was increased after treatment with ANG II, and PKC-ζ AS, but not SCR, reduced β-actin expression in the carotid artery (Fig. 7B).
Potential targeting of PKC-ζ in monocytes/macrophages cannot be excluded. Therefore, we measured the efficacy of PKC-ζ AS transfection in U-937 monocytes, cells that have been previously transfected with a high concentration of PKC-ζ AS (17). The same concentration of PKC-ζ AS (200 nM) that was effective in reducing the PKC-ζ level in VSMC and neointimal growth in injured vessels did not alter the PKC-ζ level in monocytes and in human monocytic leukemia (HL)-60 granulocytes (Fig. 7C).
Contribution of ANG II and PKC-ζ to fibrosis.
Fibrosis of injured blood vessels is caused by an increase in collagen synthesis, mainly attributable to adventitial fibroblasts (45, 58). Chronic elevation of ANG II also promotes adventitial fibrosis (6, 51). Collagen content was measured with Picrosirius red stain at day 2 (Fig. 8A) and day 7 (Fig. 8B) after injury. Neither ANG II nor PKC-ζ modulation altered adventitial collagen content in injured vessels. However, the adventitia was structured and delimited in injured and vehicle control at day 2 and disorganized and scattered with ANG II-treated arteries, similar to perivascular fibrosis, and independent of PKC-ζ. At day 7, the adventitia was again delineated. Therefore, initial exposure to ANG II did not promote more collagen synthesis but rather more disorganization of the adventitia.
This study shows for the first time that atypical PKC-ζ plays a critical role in neointimal formation and in ANG II-accelerated neointimal growth after balloon injury of the rat carotid arteries. This conclusion is based on our demonstration that exposure of balloon-injured arteries to PKC-ζ AS ODN blocked neointimal formation and inhibited the stimulatory effect of ANG II on neointimal growth. Acute or transient luminal exposure of the damaged vessel to ANG II promotes a PKC-ζ-sensitive doubling in medial cellularity, most likely through increased cell migration, which leads and correlates to an acceleration of the restenotic process. This increased medial cellularity is the foremost notable feature of ANG II treatment. ANG II also promoted a PKC-ζ-dependent increase in β-actin and VCAM levels in the vessel wall and in MCP-1+/vimentin+ cells in the media that did not involve ED1+ macrophages, proliferation, or VSMC differentiation. ANG II changed the pattern of proliferation and inflammation in the blood vessel from a perivascular to an adventitial location at day 2 after injury. Modulation of PKC-ζ in the damaged blood vessel did not alter overall cell proliferation, VSMC dedifferentiation, adventitial fibrosis, or macrophage content in the vascular wall and perivascular area. The target of PKC-ζ in the vessel wall appears to be VSMC and not leukocytes.
The target of PKC-ζ AS ODN in the vascular wall appears to be medial VSMC for the following reasons. Luminal application of PKC-ζ ODN after endothelium removal put it in close proximity with medial VSMC. PKC-ζ AS was incorporated into medial VSMC as shown by the use of a fluorescent probe and may inhibit VSMC activation or phenotypic modulation in response to apoptosis and injury (39). Upregulation of PKC-ζ elicited by ANG II in the periluminal area after balloon injury may also represent the first step in the initiation of leukocyte infiltration and neointimal growth. Proliferation and MCP-1+/VCAM+ cells in adventitia were reduced after transfection with PKC-ζ AS ODN. MCP-1 is probably expressed by remaining medial VSMC that undergo phenotypic modulation and massive apoptosis after injury (37, 39). However, MCP-1 is a nonselective marker of inflammation expressed by all cell types potentially present in the vascular wall after injury, such as VSMC, fibroblasts, and inflammatory cells (15). Decrease or loss of VSMC markers such as SM α-actin and SM22 after injury was not reversed by PKC-ζ AS, indicating that PKC-ζ does not regulate VSMC differentiation. Our data indicate that PKC-ζ AS does not block fibroblast-associated events in the adventitia, such as fibrosis. In addition, transfection of monocytes or granulocytes with 200 nM PKC-ζ AS was not successful, suggesting that the initial target of the ODN at this low concentration was not leukocytes but rather VSMC.
Decreased PKC-ζ levels after injury are probably due to an increase in PKC-ζ degradation after its activation. Indeed, several studies have shown that PKC-ζ activity is downregulated through proteolysis after phosphorylation at Thr410 and autophosphorylation at Thr560 (2, 25, 48). In addition, PKC-ζ is activated by caspases releasing the COOH-terminal catalytic domain during apoptosis and polyubiquitinated and degraded by the proteasome pathway (46). Therefore, in the carotid injury model, PKC-ζ AS blocks PKC-ζ neosynthesis after activation or apoptosis. In addition, a recent work (29) shows that aortic endothelial PKC-ζ activity correlates with increased susceptibility to atherosclerosis. Basically, PKC-ζ was regulated through a degradative pathway in areas not susceptible to atherosclerosis, whereas PKC-ζ was active in areas susceptible to atherosclerosis in the aorta. Another possibility involves medial cell apoptosis observed immediately after vascular injury (37, 39). In this case, full-length PKC-ζ level decreases in response to apoptosis, and ANG II increases PKC-ζ levels by promoting infiltration of inflammatory and noninflammatory cells from the perivascular area to the adventitia and media. It should be noted that PKC-ζ antibody is able to recognize the COOH-terminal catalytic domain of PKC-ζ (formed after activation/proteolysis) as measured by immunohistochemistry. Therefore, PKC-ζ detection on tissue sections may not account for full-length PKC-ζ protein, especially after activation or degradation after apoptosis. In both cases, inhibition of PKC-ζ neosynthesis with PKC-ζ AS in VSMC or inhibition of PKC-ζ in infiltrating cells contributed to the decrease in neointimal growth observed with PKC-ζ AS alone or in combination with ANG II.
Proliferation and migration of medial cells to the intima have been suggested to contribute to neointimal formation. However, ANG II does not stimulate proliferation of medial VSMC in early stages after injury (10, 52). Moreover, ANG II did not stimulate proliferation of VSMC in culture (data not shown). Therefore, ANG II does not appear to potentiate neointimal growth by directly stimulating medial VSMC proliferation, raising the possibility that ANG II may stimulate the synthesis of VSMC-derived mediators that will promote migration of one or more cell types and subsequent proliferation, leading to acceleration of neointimal growth. In support of this hypothesis, it is important to note that infiltrating leukocytes and resident VSMC have dissimilar cell cycle profiles and that maximal medial proliferation occurs at day 3 when <5% of VSMC are proliferating (18). Adventitial and perivascular cells probably contribute to the repair of the media, damaged by >70% apoptosis after injury (39), and to neointimal formation (26). Fibroblasts or other cell types probably need to be activated or stimulated for migration by upregulation of selective mediators in VSMC, as is the case for leukocytes. Our data clearly show a correlation between leukocytes and proliferating cells in the perivascular tissue. However, the distribution pattern of increased medial cellularity elicited by ANG II was distinct from that of BrdU labeling in the injured vessel. This suggests that cell migration, possibly of adventitial fibroblasts or other cell types, could be the main factor contributing to neointimal growth.
Extensive inflammation has been shown to correlate with ANG II-accelerated atherosclerosis, using a model of chronic ANG II infusion (3). Our data show for the first time that initial and transient exposure to ANG II (1 h) promotes the acceleration of neointimal growth in the carotid balloon injury model. The acceleration of neointimal growth induced by ANG II at day 7 and day 14 is reflected at the cellular level at day 2 by an increase in medial cellularity and at the molecular level by an increase in inflammatory markers (MCP-1, VCAM, and ED1) and proliferation selectively in the adventitial area. Our data show a correlation between medial cellularity at day 2 and the extent of neointimal growth at day 14 (Fig. 5B). In our study, ANG II potentiated monocyte/macrophage infiltration from the perivascular tissue to the adventitia of the injured vessel. Leukocytes are present in the adventitia of injured vessels within hours of injury (55). Moreover, proliferation of leukocytes precedes the proliferation of VSMC in the media (18). VSMC appear in the intima only at day 4 or day 5 after injury (7). ANG II, in our experimental conditions of short-term incubation, may, therefore, promote an enhanced expression of inflammatory proteins in VSMC, followed by an accelerated migration of one or more cell types that contribute to neointimal growth. Therefore, it is possible that transient exposure of injured arteries to ANG II stimulates cell migration from the media to the intima at an earlier stage than injury alone. In addition, ANG II may activate macrophages, platelets, and fibroblasts, independent of its effect on VSMC.
In several cell types, PKC-ζ is a positive regulator of NF-κB (13, 30, 41), a nuclear factor controlling the expression of a variety of proinflammatory signals, including the adhesion molecules ICAM and VCAM and the chemoattractant MCP-1 (53). Our data support the fact that PKC-ζ inhibition in medial VSMC may block the induction of proinflammatory genes in activated VSMC and consequently cell proliferation and migration in the intima. Whether PKC-ζ acts through NF-κB in VSMC is not known. Moreover, we have recently shown that PKC-ζ mediates ANG II-dependent VSMC adhesion on collagen (35), indicating a crucial role for this kinase in regulating adhesion to the extracellular matrix as well as mediating cell/cell adhesion as previously reported (20, 24). In addition, PKC-ζ also mediates ANG II-induced protein synthesis in VSMC (35), supporting its role as a key element for neosynthesis of inflammatory mediators by VSMC.
VSMC progenitors and stromal cell-derived factor (SDF)-1 contribute to neointima formation after arterial injury (59). Recently, PKC-ζ has been shown to regulate SDF-1-mediated migration of immature progenitor cells in the bone marrow (38). Moreover, overexpression of PKC-ζ in U-937 cells leads to increased directional motility to SDF-1 (38). Therefore, the role of PKC-ζ on VSMC progenitor recruitment after vascular injury may be essential for neointimal formation. ANG II accelerates neointimal growth by attracting more VSMC precursors, either circulating or resident progenitors, adventitial myofibroblasts, or surviving dedifferentiated medial VSMC, than injury alone. Macrophages present in the adventitial and perivascular tissues after injury are terminally differentiated, and their infiltration was accelerated by ANG II, independent of PKC-ζ inhibition; therefore, they are unlikely to play a significant role in ANG II-accelerated restenosis (54).
In conclusion, our study demonstrates for the first time that inhibition of PKC-ζ expression with an antisense ODN blocks neointimal formation and ANG II-accelerated restenosis. PKC-ζ plays a major role in promoting an early medial infiltration in the carotid artery that consequently controls the extent of neointimal growth. In addition, our data suggest that medial VSMC are the site of PKC-ζ AS ODN action (Fig. 9). Targeting of PKC-ζ may represent a new therapeutic strategy for attenuation of restenosis after angioplasty.
This research was supported by a Beginning Grant-In-Aid of the American Heart Association (Southeast Affiliates, to J.-H. Parmentier), a Junior Faculty Grant from American Diabetes Association (to C. Zhang), and a National Heart, Lung, and Blood Institute Grant HL-79109-02 (to K. U. Malik).
We thank Dr. Lauren Cagen for editorial comments.
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 © 2006 by the American Physiological Society