OUP user menu

Adventitial application of the NADPH oxidase inhibitor apocynin in vivo reduces neointima formation and endothelial dysfunction in rabbits

Elsa C. Chan, Srinivasa R. Datla, Rodney Dilley, Haruyo Hickey, Grant R. Drummond, Gregory J. Dusting
DOI: http://dx.doi.org/10.1016/j.cardiores.2007.06.005 710-718 First published online: 1 February 2008


Objective Reactive oxygen species including superoxide have been shown to promote atherogenesis. We previously showed that a major source of superoxide, the NADPH oxidase system, is upregulated in the intima and adventitia during remodelling induced by periarterial collars in rabbits. We have now examined the action of the NADPH oxidase inhibitor apocynin, given via the adventitia, on the neointima formation and endothelial function in this model.

Methods Perivascular collars were implanted around the common carotid arteries of male NZW rabbits for 14 days to induce intimal thickening. The periarterial space of one collar was filled with apocynin (1 mM) while the contralateral collar with the vehicle (0.1% DMSO).

Results After 14 days, local treatment with apocynin via the adventitia, reduced superoxide generation. In addition, apocynin significantly reduced neointima formation and proliferation of cells in both the neointima and adventitia. Moreover, NO-dependent vasorelaxation to acetylcholine, which is normally impaired in collared arteries, was improved, and apocynin suppressed the endothelial expression of intracellular adhesion molecule-1 and vascular cell adhesion molecule-1.

Conclusions NADPH oxidase is implicated in vascular remodelling and superoxide-stimulated cell proliferation in the neointima contributes to intimal hyperplasia in this collar model. Targeting NADPH oxidase via adventitial drug delivery not only reduces superoxide generation, but also normalises endothelial cell function. Targeting the primary source of NADPH oxidase-derived superoxide is an effective approach to prevent deleterious arterial remodelling, providing a rationale for designing more efficacious and selective inhibitors of vascular NADPH oxidase as potential therapeutics for human vascular disease.

  • Adventitial drug delivery
  • Apocynin
  • Cell proliferation
  • Endothelial dysfunction
  • NADH oxidase
  • Neointima formation
  • Perivascular collar-induced remodelling

Time for primary review 32 days

1 Introduction

Reactive oxygen species (ROS) are thought to play an important role in atherogenesis. We have demonstrated that the Nox2 catalytic component of NADPH oxidase is upregulated in a non-hyperlipidaemic rabbit model of early stage arterial remodelling [1]. This enzyme system is composed of two cytosolic subunits p47phox and p67phox, a cell membrane-bound cytochrome b558 which consists of gp91phox (or its homologues) and p22phox, and a small G protein rac. Upon assembly of these subunits in the membrane, this enzyme generates superoxide anion by one-electron reduction of oxygen via its gp91 subunit using NADPH as the electron donor [2]. NADPH oxidases appear to be the major sources of superoxide production in vascular tissues (reviewed by Griendling et al. [3] and Jiang et al. [2]). Association has been made between elevated superoxide levels and neointima formation. This is apparent in models of restenosis and balloon catheter injury [4–7] or in atherosclerotic lesions in Watanabe heritable hyperlipidaemic and cholesterol-fed rabbits [8,9].

The inflammatory component of the intimal thickening in human early atheroma is mimicked by implanting perivascular collars in rabbit carotid arteries [10–12]. Such characteristics include endothelial dysfunction, macrophage infiltration and deposition of collagen and fibronectin within the intima, and accumulation of cholesteryl esters in the artery wall, despite normocholesterolaemia [12]. In addition, cells within the intima have a similar morphology to “synthetic” vascular smooth muscle cells [12], confirming that proliferation and migration of vascular smooth muscle cells have key roles in intimal hyperplasia. Although the morphological changes in the collar-induced neointima have been well characterised [10–12], the stimuli and underlying molecular mechanisms of the neointimal hyperplasia are not fully understood. We have previously identified that Nox2 (formerly known as gp91 phox) expression is upregulated and is associated with elevated endothelial and adventitial NADPH oxidase-dependent superoxide production in this model of vascular remodelling [1], leading to endothelial nitric oxide dysfunction [11]. Interestingly, Pagano and colleagues [13] showed that gene transfer of an NADPH inhibitor peptide via the adventitia, reduced neointima formation after balloon injury to the intima in rats. More recently, adventitial delivery of a dominant viral vector specifically suppressing the NADPH oxidase p67phox subunit was shown to attenuate the intimal hyperplasia in balloon-injured carotid artery in rats [14]. Therefore, we set out to investigate the effect of reducing NADPH oxidase-derived superoxide on arterial remodelling in our collar model by applying an NADPH oxidase inhibitor to the adventitia. One of the advantages of the collar model is that it allows the effects of drugs to be examined locally, without the complication of systemic actions. Apocynin given this way probably penetrates all layers of the vessel wall and we show that the inhibitor suppressed superoxide generation and artery remodelling as well as cell proliferation in both the neointima and adventitia, and restored endothelial function in collared arteries.

2 Methods

2.1 Rabbits

Male New Zealand White rabbits weighing 3-4 kg (Nanowie Small Animal Production Unit, Modewarre, Victoria, Australia) were maintained on a normal chow diet throughout the experiments. The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

2.2 Collar implantation and local drug treatment in vivo of carotid artery

Rabbits were anaesthetised with propofol (5 mg/kg, IV) followed by ketamine/xylazine (50/10 mg/kg, IM). Left and right common carotid arteries were exposed surgically and cleared of connective tissue along a 30-mm length. Hollow, nonocclusive silastic collars (length 20 mm; internal diameter along bore 4 mm; internal diameter at ends 1 mm) were then placed around each artery and held in place by a nylon sleeve. The periarterial space of one collar was filled with apocynin (1 mM) while the contralateral collar with the vehicle DMSO (0.1%), referred as “vehicle-collared artery”. The apocynin filled collar (apocynin-collared artery) was then attached to a mini-osmotic pump (ALZET®, Model 2001, Durect Corporation, Cupertino, U.S.A; infusion rate of 1 μl hr-1) via a polyethylene catheter delivering apocynin (1 mM) to ensure a constant presence of the drug solution in the collar. Muscles, fat and skin layers were sutured and wound dressed with antibiotic. Rabbits were allowed to recover for 14 days following apocynin treatment.

2.3 Tissue harvesting

Rabbits were heparinized (1000 IU, IV) before euthanasia with sodium pentobarbitone (90 mg kg-1, IV). Carotid arteries were excised and placed in ice-cold Krebs-HEPES buffer (pH 7.4). Arteries were cleaned of fat and connective tissues following collar removal. Three ring segments (∼3 mm) were cut from the area enclosed by the collar and from the proximal or distal non-collared segment of artery for use as within-animal controls. An additional ring segment (∼ 1 mm) was taken at the centre of each region for morphological analysis. One section from each non-collared and collared segments was snap-frozen in Tissue-Tek OCT (Sakura) fixative. Frozen tissues were sectioned in 10-μm slices and fixed with acetone at -20 °C for 15 min to detect adhesion proteins and proliferating cells using immunofluorescence and immunohistochemistry respectively.

2.4 Measurement of neointima thickening

Following overnight fixation in 4% paraformaldehyde at 4°C, carotid artery segments were dehydrated in ethanol and embedded in paraffin. Transverse sections (4 μm) were stained with haematoxylin and eosin. Using a MicroComputer Imaging Device (version 3.0, Imaging Research Inc), the cross-sectional areas of the intimal and medial regions were measured in micrometers squared and expressed as an intima-media ratio (IMR) to quantify neointima formation.

2.5 Ex vivo (lucigenin-enhanced chemiluminescence) and in situ (dihydroethidium) detection of superoxide in carotid arteries

Levels of superoxide in non-collared and collared arteries were quantified by lucigenin (5 μM)-enhanced chemiluminescence in the presence of NADPH (100 μM) as previously validated [1,15]. To eliminate the influence of endogenous Cu/Zn SOD activity, all aortic segments were treated with diethyldithiocarbamic acid (DETCA, 3 mM) for 45 min prior to chemiluminescence assay. Dihydroethidium (5 μM) was used as previously described to detect superoxide in frozen sections (10 μm, n=4) of non-collared and collared arteries [1,8,9]. Sections preincubated with a SOD mimetic MnTmPyP (25 μM) virtually eliminated fluorescence and were used as background control (data not shown). The sections were viewed under the fluorescence microscope (excitation at 546/12 nm and detection at 590 nm) and the images were taken at fixed settings for all sections. The red fluorescent density was measured with Image J software [16] and was normalised to the area of artery segment.

To investigate whether the vehicle DMSO (0.1%) might have affected superoxide generation in the collar, rabbit thoracic aortic segments (n=4) were incubated with and without DMSO (0.3%) for 30 min at 37 °C prior to superoxide quantification using lucigenin (5 μM)-enhanced chemiluminescence in the presence of NADPH (100 μM).

2.6 Expression of adhesion proteins

To detect vascular adhesion proteins, acetone fixed sections were incubated with murine monoclonal antibodies against rabbit vascular cell adhesion molecule-1 (VCAM-1) and intracellular adhesion molecule-1 (ICAM-1, both antibodies are gifts from Dr M. Cybulsky, University of Toronto) for 2 h at room temperature, followed by 1-h incubation with Cy3 conjugated goat anti-mouse antibody at room temperature. Sections were mounted and visualised using fluorescence microscopy. Total area of antigen-specific fluorescence staining per section was quantified using Scion Image (Beta 4.02) imaging software. Positively stained areas were selected and measured using program-specific tools. Two replicate measurements were averaged for each tissue type per rabbit (n=6) to obtain data for statistical analysis.

2.7 Cell proliferation

To determine the extent of cell proliferation in non-collared and treated collared arteries, acetone fixed sections were incubated with mouse antibody against human Ki67 protein (Dako) overnight at 4 °C. Sections were then incubated with biotinylated anti-mouse immunoglobulin for 30 min and streptavidin with peroxidase label for 30 min. The resulting peroxidase reaction was revealed by diaminobenzidine. Sections were then counter stained with haematoxylin and mounted. Ten to 16 images from the neointimal, medial and adventitial layers were randomly selected to calculate total cell number with a minimum of 250 cells. The proliferation index (%) was then expressed as the number of Ki67 positive cells within the total number of counted cells. Slides were rinsed in PBS (pH 7.4) after each incubation during the staining process. Negative controls were performed using a mouse immunoglobulin G (IgG) for both immunofluorescence and immunohistochemistry.

2.8 Assessment of endothelial nitric oxide function

Vasorelaxant responses to acetylcholine (0.001 to 100 μM) and sodium nitroprusside (0.001 to 30 μM) in isolated non-collared and collared artery rings in standard organ baths were determined as previously described [1]. Relaxations were expressed as percent relaxation of serotonin-induced precontraction. Non-linear regression (Prism version 4.0) was used to determine pEC50 and maximum responses (Rmax) to acetylcholine and sodium nitroprusside in each artery ring.

2.9 Data analysis

All results are expressed as mean±standard error of the mean (sem) from the number of performed experiments (n). Statistical comparisons were made by one-way ANOVA with Bonferroni or Tukey corrections where appropriate (Graph Pad Prism version 4.0). A value of P<0.05 was considered significant.

3 Results

3.1 Superoxide production in apocynin-collared arteries

Superoxide production in artery segments (n=5-6) was detected by lucigenin chemiluminescence. Collaring the carotid arteries for 14 days significantly enhanced superoxide generation by 7-fold when compared to non-collared segments (Fig. 1A). Local treatment of collared carotid arteries with apocynin (1 mM) in vivo significantly blunted superoxide generation (Fig. 1A).

Fig. 1

Effect of apocynin (1 mM) on superoxide generation in collared carotid arteries, detected by lucigenin (5 μM)-enhanced chemiluminescence in the presence of NADPH (100 μM, A) and dihydroethidium (5 μM, B). Fourteen-day vehicle-treated periarterial collar placement significantly increased superoxide generation in comparison to those of non-collared segments (*P<0.05). Apocynin effectively reduced the collar-induced superoxide production (*P<0.05). Values (mean+sem from 5 to 6 experiments) are expressed as multiples of increase in superoxide generation compared to those in non-collared arteries detected by lucigenin-enhanced chemiluminescence (A). Dihydroethidium staining of superoxide in collared and non-collared arteries (B) is expressed as the ratio of the density of red fluorescence emitted from the superoxide reactive dihydroethidium per area of artery segment (mean+sem from 4 experiments).

Similar to the above, dihydroethidium (5 μM) fluorescence was significantly increased in collared arteries when compared to the controls (Fig. 1B, n=4) and was predominantly localised in the endothelial layer and the adventitia (Fig. 2B). Apocynin treatment of the collared arteries significantly reduced dihydroethidium fluorescence (Fig. 1B) showing little red fluorescence in the endothelial and adventitial layers (Fig. 2C). Minimal dihydroethdium fluorescence was found in the non-collared artery (Figs. 1B and 2A).

Fig. 2

In situ localisation of superoxide production in transverse sections (10 μm) of rabbit carotid artery. Non-collared (A), vehicle-collared (B) and apocynin-collared (C) artery segments were treated with dihydroethidium (5 μM) and DETCA (3 mM) before visualisation under a fluorescence microscope. The dihydroethidium fluorescence is predominantly found in the endothelial layer and the adventitia in the vehicle-collared artery section (A), and is suppressed by the local in vivo treatment with apocynin (C). Minimal fluorescence is found in the non-collared artery (A). Scale bar represents 50 μm. Each image is a representative from 4 rabbits.

To investigate whether the vehicle of apocynin DMSO (0.1%) might have affected the superoxide generation in collared arteries, rabbit thoracic aorta was incubated with up to 0.3% of DMSO prior to and throughout superoxide detection by lucigenin-enhanced chemiluminescence. DMSO (0.3%) did not affect the NADPH (100 μM) -driven superoxide production in vessels when compared to untreated segments (1277±153 CPS per mg in DMSO treated vessels versus 1268±287 CPS per mg in untreated vessels, n=4).

3.2 Effect of apocynin on neointima formation induced by perivascular collars

The medial area in both vehicle- and apocynin-collared arteries was significantly reduced when compared to the non-collared segment (Table 1). In collared arteries, apocynin treatment did not affect the medial changes but it significantly suppressed the neointimal growth showing a reduction in both intimal area and IMR when compared to vehicle treatment (Table 1). The non-collared segment remained unaffected (Table 1).

View this table:
Table 1

Effect of local treatment with apocynin (1 mM) in vivo on neointima formation in carotid arteries

TreatmentnMedia (mm2×105)Intima (mm2×105)IMR
Vehicle collar104.79±0.21*0.76±0.13*0.16±0.02
Apocynin collar104.70±0.05*0.47±0.05†0.10±0.01†
  • *Significantly different from non-collared artery (P<0.05); † significantly different from vehicle collared artery; undefined due to the lack of intimal growth. Values are expressed as the mean±sem from 10-15 (n) experiments.

3.3 Effect of apocynin on cell proliferation in collared arteries

To examine the contribution of cell proliferation in artery remodelling in our collar model (n=4), Ki67-specific cells in the intima, media and adventitia were counted in total number of cells in each of the three layers (Fig. 3). When compared to the non-collared artery, the adventitia and neointima in the vehicle-collared artery both showed a significant increase in the proliferation index (Fig. 4). Apocynin treatment effectively suppressed these neointimal and adventitial changes and reduced the adventitial proliferation index to control level (Fig. 4). Interestingly after 14 days of collar placement smooth muscle cell proliferation indices were comparable in non-collared and vehicle- and apocynin-collared arteries (Fig. 4). The specificity of Ki67 antibody was confirmed and there was no positive staining in artery sections incubated with the negative control mouse IgG (data not shown).

Fig. 4

Effect of apocynin (1 mM) on cell proliferation indices in neointima (A), media (B) and adventitia (C) in vehicle-collared (open bar) and apocynin-collared (grey bar) carotid arteries. There were significant increases in cell proliferation indices in the neointima and adventitia in vehicle-collared artery (*P<0.05) when compared to non-collared artery (black bar). Apocynin effectively suppressed these increases (*P<0.05) in the collared arteries. Values (mean+sem from 4 experiments) are expressed as the percentage of cells expressing Ki67 protein within the total number of counted cells. †P<0.05 when compared to non-collared artery.

Fig. 3

Effect of apocynin (1 mM) on cell proliferation in neointima, media and adventitia in non-collared (A), vehicle-collared (B) and apocynin-collared (C) carotid arteries. Proliferating cells expressing Ki67 are shown in dark brown and some are identified by black arrows. There were an increased number of Ki67 positive cells in the neointima and adventitial in the vehicle-collared artery while no proliferating cells were found in the non-collared section. Apocynin treatment suppressed the number of proliferating cells in the neointima and adventitia. The media in non-collared and vehicle-collared and apocynin-collared arteries showed minimal Ki67 positive cells. Each image is a representative of 10-16 images taken in non-collared, vehicle-collared and apocynin-collared arteries from 4 rabbits. Scale bar represents 25 μm.

3.4 Effect of apocynin on expression of vascular adhesion molecules

Intimal thickening was also associated with ROS-dependent accumulation of adhesion proteins. Therefore the ability of apocynin to inhibit the expression of vascular VCAM-1 and ICAM-1 was also investigated (n=6). Immunofluorescence detection revealed an upregulation of vascular adhesion molecule expression in sections of collared arteries when compared to non-collared segment (Fig. 5). Adhesion molecule expression within the vascular wall was predominantly localised to the endothelium (Fig. 5). However, in collared vessels, positive immunoreactivity was also observed around the adventitia, probably reflecting development of vasa vasorum. The expression of adhesion molecules ICAM-1 and VCAM-1 was assessed by specific immunofluorescent signals. Perivascular collar treatment significantly increased ICAM-1 and VCAM-1 immunofluorescence (Table 2). Consistent with its preventative effects on neointima formation, intimal and adventitial cell proliferation and superoxide generation, apocynin effectively reduced VCAM-1 staining compared to those of collared rings (Table 2), and tended to reduce ICAM-1 expression (Table 2). Sections incubated with mouse IgG showed no expression of ICAM-1 or VCAM-1 (data not shown), confirming the specificity of both antibodies.

Fig. 5

Immunofluorescence localisation of VCAM-1 expression in transverse sections of control (A), vehicle-collared (B) and apocynin-collared (C) artery segments. Positioning of collar for 14 days increased the endothelial expression of VCAM-1 (B), which was effectively suppressed by local treatment with apocynin in vivo (C). Sections were incubated with specific anti-rabbit VCAM-1 antibody, followed by a Cy3-conjugated antibody prior to examination under a fluorescence microscope (10X magnification). Images are representatives of sections from 6 rabbits. Scale bar represents 100 μm.

View this table:
Table 2

Effect of local treatment with apocynin (1 mM) in vivo on expression of adhesion molecules in carotid arteries

TreatmentnVCAM-1 (mm2×104)ICAM-1 (mm2×104)
Vehicle collar617±7*59±24*
Apocynin collar66±2†16±1*
  • *Significantly different from non-collared artery (P<0.05); † significantly different from vehicle-collared artery (P<0.05). Positively stained areas were selected and measured using Scion Image program-specific tools. Two replicate measurements were averaged for each tissue type per rabbit. Values are expressed as the mean±sem from 6 (n) experiments.

3.5 Effect of apocynin on endothelium-dependent vasorelaxations in collared arteries

We have previously showed that the impairment of endothelium-dependent vasorelaxation in collared arteries was attributable to enhanced superoxide production [1]. Therefore endothelium-dependent and -independent vasorelaxant responses were compared in non-collared and vehicle- and apocynin-collared arteries ex vivo. Acetylcholine caused concentration-dependent vasorelaxation in all isolated arteries (Fig. 6). The sensitivity (pEC50) and the maximum response (Rmax) to acetylcholine were both significantly reduced in collared arteries (Fig. 6 and Table 3).

Fig. 6

Effect of apocynin (1 mM) on relaxation to acetylcholine (ACh) in isolated carotid arteries. Responses to ACh (vehicle treatment in open triangles, pEC50 of 6.7±0.2; apocynin treatment in filled triangles, pEC50 of 6.8±0.1) were reduced in collared arteries when compared to non-collared segments (filled squares; pEC50 of 7.6±0.1; P<0.05). Apocynin restored the maximum relaxation to those of non-collared segments (*P<0.05). Values (mean±sem from 6 to 12 experiments) are expressed as percent relaxation of the 5-HT-induced precontraction.

View this table:
Table 3

Effect of local treatments with apocynin (1 mM) in vivo on responses to acetylcholine and sodium nitroprusside

TreatmentnAcetylcholineSodium Nitroprusside
pEC50Rmax (%)pEC50Rmax (%)
Vehicle collar66.7±0.2*70±9*7.8±0.2101±1
Apocynin collar66.8±0.1*91±4†7.7±0.1100±1
  • Rmax was expressed as % of 5-HT-induced precontraction; *significantly different from non-collared rings, P<0.05; †significant different from vehicle-collared arteries, P<0.05; Values are expressed as the mean±sem from 6-12 (n) experiments.

Local treatment with apocynin in vivo restored the maximum response to acetylcholine to that of non-collared arteries without affecting their sensitivity (Fig. 6). In contrast to the impaired responses to acetylcholine, the sensitivity and maximum responses to the nitric oxide donor, sodium nitroprusside, in collared arteries were similar to those of non-collared arteries and were not affected by apocynin treatment (Table 3).

4 Discussion

This is the first study to show that suppressing NADPH oxidase activity in vivo in the rabbit collar model of arterial remodelling reduces the intimal thickening and cell proliferation in both neointima and adventitia. In addition, the inhibitor suppressed expression of adhesion molecules in the intima and prevented endothelial dysfunction. Therefore, direct inhibition of the source of superoxide generation with apocynin may well be a useful approach to reduce the remodelling associated with the early stages of post-angioplasty remodelling or atherosclerosis.

We previously showed that the arterial remodelling induced by implanting perivascular collars on rabbit carotid arteries causes an upregulation of the NADPH oxidase catalytic subunit Nox2 even though this remodelling occurs in a normolipidaemic environment without major participation of macrophages [1]. We now demonstrate that local administration of an NADPH oxidase inhibitor apocynin in vivo to the collared arteries attenuates superoxide production and prevents vascular injury associated with this remodelling. Apocynin is an inhibitor of NADPH oxidase and has been shown to prevent the translocation of the cytosolic subunits to the membrane bound catalytic subunit [17], thereby inhibiting the formation of an active NADPH oxidase complex. Apocynin treatment of the collared arteries did not affect Nox2 mRNA expression (data not shown), thus indicating that apocynin reduced superoxide generation in the collared arteries by suppressing the NADPH oxidase activity rather than affecting its gene transcription. Moreover, medium term inhibition of NADPH oxidase activity does not lead to upregulation of this major catalytic subunit.

Apocynin was given to the collared arteries in the presence of 0.1% DMSO as the vehicle in the current study. Although DMSO is known to scavenge free radicals, the administered DMSO is minimal and it appeared not to have any effect on superoxide generation. Furthermore rabbit aortic segments preincubated with and without a higher concentration of DMSO (0.3%) produced similar levels of superoxide upon NADPH stimulation, thus eliminating the possibility of a DMSO contribution to the observed effect of apocynin in collared arteries. In this model, apocynin was administered into the collar space enclosing the carotid artery and it is not anticipated to have entered the systemic circulation via diffusion in significant amounts. This is confirmed by the fact that the level of dihydroethidium fluorescence in the non-collared artery proximal or distal to the apocynin-collared section is similar to their contralateral non-collared segments (data not shown).

Apocynin has been demonstrated to reverse endothelial NO dysfunction in animals or humans subjected to oxidative stress [18]. Several studies have reported the protective effect of in vivo treatment with apocynin in experimental vascular injury models associated with ROS overproduction. Beswick et al [19] and Gosh et al. [20] showed that extended oral treatment of apocynin not only reduces superoxide generation in arteries isolated from deoxycorticosterone acetate salt (DOCA)-induced hypertensive rats, but also reduces blood pressure associated with hypertension [19,20]. Recently apocynin given orally was also found to prevent and reverse the adrenocorticotropic hormone (ACTH)-induced hypertension in rats [21]. Thus apocynin appears to be an effective NADPH oxidase inhibitor in vivo and acts to prevent vascular damage associated with oxidative stress.

NADPH oxidase-derived ROS including superoxide are known to promote atherosclerosis through stimulation of proatherogenic transcription factors such as NFkB [22,23]. Recently, adventitial delivery of a viral vector suppressing the expression of the NADPH oxidase subunit p67phox reduced the intimal thickness and area of carotid artery in rats undergoing inflated balloon injury [14]. The dominant-negative p67phox viral vector did not affect the medial area and thickness in the injured arteries in the same study [14]. Similarly apocynin administered during intimal thickening in our collar model reduced the neointima formation by suppressing the neointimal area without affecting the medial atrophy. The observed medial atrophy in the collared arteries appears to result from imbalance of the migration and proliferation of vascular smooth muscle cells through the internal elastic lamina into the intima [10]. This tends to occur during the early phase after applying the collar [12] since the degree of medial cell proliferation is comparable between the non-collared and collared arteries harvested on day 14. Given that the apparent enhanced cell proliferation in the neointima and adventitia in the collared arteries is suppressed by apocynin, and superoxide generation is enhanced in the endothelial and adventitial layers [1], the neointimal and adventitial cell proliferation appears to follow from generation of NADPH oxidase-derived superoxide. Thus, as found in other models, NADPH oxidase-derived superoxide is strongly implicated in the proliferation of vascular cells in the artery wall and particularly within the neointima [6]. Consistent with this, mice overexpressing the NADPH oxidase subunit p22 phox developed a substantial atheroma [24] and Jacobson et al [25] demonstrated that intraperitoneal delivery of the chimeric NADPH oxidase inhibitor peptide gp91ds-tat reduced neointimal hyperplasia in the injured carotid artery.

We identified the neointima and adventitial layers as sites of superoxide generation in our collar model [1]. Apart from its direct atherogenic activity, superoxide inactivates and reduces vasoprotective NO bioavailability through the production of the proatherogenic ROS peroxynitrite [26]. Hence these events would all promote lesion development because NO inhibits smooth muscle cell migration and proliferation [27] while peroxynitrite induces endothelial cell apoptosis [28] and uncouples endothelial NO synthase by oxidizing the essential cofactor tetrahydrobiopterin [15]. Endothelial dysfunction in these studies was modest compared with previous studies [11,29], but local administration of apocynin in vivo improved endothelial function by restoring the maximum relaxation to acetylcholine in the collared arteries. This is possibly due to the fact that apocynin, by suppressing NADPH oxidase, indirectly limited the interaction between superoxide and NO upon acetylcholine stimulation, thereby preventing the impairment of responses to acetylcholine in the collared vessels. Although the level of NO has not been measured here, the impaired endothelium-dependent relaxation in the collared arteries suggests that level of bioavailable NO has been compromised. Supporting this interpretation, Hamilton et al. [18] showed that apocynin (30-300 μM) increased the NO level in cultured human sapheous vein endothelial cells, and dose-dependently induced vasorelaxation of precontracted human isolated internal mammary arteries, which was abrogated by NG-nitro-L-arginine methyl ester. Clearly, apocynin inhibits the generation of NADPH-derived superoxide and prevents the damaging interaction between NO and superoxide to form more potent ROS, thereby maintaining endothelial function despite the stimulus for arterial remodelling. Furthermore, endothelial dysfunction in the early stages of atherosclerosis would have several detrimental effects such as increased expression of cell adhesion molecules [30], thereby promoting atherogenesis. Here we also demonstrated the effectiveness of apocynin in reducing the endothelial expression of VCAM-1 and ICAM-1 in collared arteries. These findings are therefore in agreement with the observation that Rac1 and superoxide are essential for Tumour Necrosis Factor (TNF)α-stimulated VCAM-1 and ICAM-1 expression in human aortic endothelial cells [31]. TNFα is a major proinflammatory factor involved in the development of vascular inflammation and atherosclerosis [32].

Apocynin in this study has been delivered via the adventitia. Although it clearly has powerful actions in the adventitia, exemplified by suppression of ROS production and suppression of adventitial cell proliferation, the drug also has actions throughout the vessel wall. Clearly, improvement of endothelial function, reduced adhesion molecule expression, suppression of intimal ROS and reducing neointimal cell proliferation and thickening indicate that apocynin either physically penetrates to these cells, or has an indirect influence on the neointima via release of paracrine factors from adventitial cells. Most likely both mechanisms are operating. However it is clear from this and previous work [14] that NADPH oxidase expressed in the adventitia has implications for arterial function and remodelling.

In conclusion, we found that activation of NADPH oxidase and the subsequent stimulation of cell proliferation in neointima and adventitia play crucial roles in vascular remodelling. Targeting the primary source of NADPH oxidase-derived superoxide is an effective approach to reduce detrimental arterial remodelling such as that in the early stages of atherosclerosis, hence providing a rationale for designing more efficacious inhibitors of vascular NADPH oxidase as potential therapeutics for human vascular disease.


This work was supported by an Institute Block Grant (NHMRC No.983001) and project grant (30013) from the National Health and Medical Research Council of Australia. GJD is a Principal Research Fellow of NHMRC (40003). We thank Meetali Kanagasundaram for her excellent technical assistance.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
View Abstract