Bis (Aspirinato) Zinc (II) Complex Successfully Inhibits Carotid Arterial Neointima Formation after Balloon-injury in Rats

Abstract

Purpose Neointima formation following angioplasty is a se- rious consequence of endothelial damage in arteries. Inflam- matory mediators and lack of endothelial regulatory mecha- nisms lead to migration and proliferation of smooth-muscle cells and thus to restenosis. This study examines the effect of the novel bis (aspirinato) zinc (II) complex on neointima formation in a rat model of carotid balloon-injury.

Methods Rats underwent balloon-injury of the right common carotid artery, then received PEG400 vehicle (untreated- group), acetylsalicylic-acid (ASA-group), zinc-chloride (Zn-group) and bis (aspirinato) zinc (II) complex (Zn(ASA) 2-group) orally for 18 consecutive days. From harvested ca- rotid arteries, histology, immunohistochemistry and mRNA expression analysis were performed.

Results Compared to the untreated-group, Zn (ASA) 2-treatment significantly lowered stenosis ratio (54.0±5.8 % to 25.5± 3.9 %) and reduced neointima/media ratio (1.5±0.2 to 0.5± 0.1). Significantly higher alpha smooth muscle actin mRNA and protein expression were measured after Zn (ASA)2 and Zn-treatment in comparison with the untreated and ASA-groups while the expression of matrix-metalloproteinase-9 was significantly higher in these groups compared to Zn (ASA)2. The presence of collagen in media was significantly decreased in all treated groups. mRNA expressions of nuclear factor kappa-b, transforming growth-factor-β and proliferat- ing cell nuclear antigen were significantly down-regulated, whereas a20 was up-regulated by Zn (ASA)2 treatment com- pared to the untreated and ASA-groups.

Conclusion This study proves the effectivity of the novel bis (aspirinato) zinc complex in reducing neointima formation and restenosis after balloon-injury and supports the hypothesis that inhibition of smooth-muscle transformation/proliferation plays a key role in the prevention of restenosis.

Keywords : Neointima . Restenosis . Bis (aspirinato) zinc . Balloon-injury

Introduction

The endothelial layer is mechanically compromised during percutaneous transluminal coronary angioplasty (PTCA) irre- spective of whether a stent is implanted or not. Almost 30 % of the patients who undergo PTCA develop restenosis within 6 months in the injured arterial segment, which necessitates further interventions in the long term follow up in order to maintain blood supply of the underlying myocardium [1]. The rate of restenosis following successful PTCA doubled in diabetes and metabolic syndromes.

The pathophysiology of the restenosis phenomenon is widely studied but is not yet completely understood and is even less auspiciously preventable. Neointimal proliferation as the main cause of restenosis is strictly related to the damage that the endothelium suffers from during interventions. An absent or dysfunctional endothelium does not hinder the pro- motion of vascular smooth muscle cell (VSMC) migration and proliferation [2], which is further provoked by paracrine me- diators released from activated platelets (platelet derived growth factor, insulin-like growth factor, transforming growth factor (TGF)-ß, serotonin, adenosin-diphosphate, thrombox- ane A2) on the exposed subendothelial surface, and from leukocytes (osteopontin, thrombospondin, matrix metalloproteases (MMPs)) invading the intimal layers [3–5]. The magnitude of leukocyte activation after angioplasty is known to correlate with the late restenosis [6] along with late clinical events [7]. Uneliminated reactive oxygen species (ROS) produced by NADPH oxidases of inflammatory cells and even in VSMCs also act as second messengers, mediating VSMC growth and migration to the intima [8]. By the stimuli of these factors, differentiated VSMCs with remarkably high plasticity concomitantly transform from a contractile (high aSMA, calponin presence) to a proliferative phenotype (high PCNA and MMP presence). Elucidating the mechanisms which are controlling VSMC transformation and prolif- eration is therefore a keystone in developing therapies against neointima−formation [9].

Acetylsalicylic acid is a traditional non–steroid anti–in- flammatory drug in high dose and a successful anti–aggrega- tional drug after PTCA in low dose. Although in the experi- mental model of restenosis it proved to be ineffective against neointima formation by itself [10], it may indirectly contribute to the prevention of restenosis by hindering platelet activation and accumulation [5].

The trace element zinc (involved in various cellular func- tions as co–factor of hundreds of enzymes), on the other hand, is known to possess anti–inflammatory and anti-proliferative effects. [11–13], and successfully decreased neointima forma- tion in an experimental model of carotis injury [14]. However, zinc and acetylsalicylic acid in the form of a bis (aspirinato) zinc (II) complex (Zn-aspirinate) have been shown to have three times higher anti-inflammatory potency than the physi- cal mixture of zinc and aspirinate [15], and to improve type 2 diabetes in animal model. [16]

Based on the concept that early restoration of functional endothelium results in lower restenosis rate [17] by reducing inflammatory and oxidative processes as well as the prolifer- ative drive of VSMCs, we hypothetized that effective inhibi- tion of inflammation with the novel bis (aspirinato) zinc (II) complex (or Zn-aspirinate) might hinder the aforementioned mechanisms till the functional endothelium recovers.

Methods

Animals

For the experiments young male Sprague–Dawley rats (300 g body weight) were used. Animals were obtained from Charles River, Sulzfeld, Germany. They were housed in a room at a constant temperature of 20±2 °C with 12-h light/dark cycles and were fed a standard dry laboratory rat diet and water ad libitum. All procedures concerning animals were conform the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources and pub- lished by the National Institutes of Health (NIH Publication No. 86–23, revised 1996) and the German animal protection code. Approval was also granted by the local ethics review board (G-135/13).

Experimental Groups

Rats were randomly assigned into four experimental groups (12 animals per group) and underwent balloon-injury of the right carotid arteries as described below. The animals received daily oral medication of 1) ASA group: acetylsalicylic-acid (90 mg/kg body weight diluted in water), 2) Zn group: zinc- chloride (31 mg/kg body weight with 15 mg/kg body weight zinc content diluted in water), 3) Zn (ASA)2 group: bis (aspirinato) zinc (II) complex (105 mg/kg with 15 mg/kg body weight Zn and 90 mg/kg ASA content diluted in polyethylene-glycole 400) and 4) injured control group (1 ml/kg body weight polyethylene-glycole 400 vehicle). After 18 days of treatment carotid arteries were harvested for further analysis. Absolute control samples (unoperated and untreated) were picked from the left carotids of the injured control group. The treatment caused no difference in the body weight of the animals.

Balloon Injury of the Carotid Arteries

A balloon injury model of rat carotid artery was performed as described in the literature [14]. Rats were anaesthetised with a mixture of ketamine (100 mg/kg body weight) and xylazin (5 mg/kg body weight). The central cervical skin was shaved and by using a dissecting microscope the right common carotid artery with the bifurcation was isolated from a median cervical incision. The internal carotid artery was clamped along with the first posterior branch of the external carotid as they run close to each other. The external carotid was clamped proximally from its second branch, approximately 3–4 mm far from the first branch. The common carotid artery was clamped close to the crotch of the subclavian artery. On the 3–4 mm segment of the external carotid a longitudinal incision was made, and a 2 F Avion Plus (1.25 mm) balloon- catheter was inserted 2–2.5 cm deep into the artery. The balloon was inflated with 7 bar nominal pressure and slightly twirled while being passed 3 times along the common carotid artery. After deflating the balloon, the catheter was withdrawn and the longitudinal incision on the external carotid was closed with a 9–0 thread. The neck incision was closed with 4–0 running suture. Animals received 0.05 mg/kg body weight buprenorphine-hydrochloride for analgesic 2 times in the next 24 h.

Excision of the Common Carotid Arteries

Eighteen days after the arteriotomy and balloon-injury, rats were anaesthetised with isofluran. The depth of narcosis was checked with the interfinger reflex. Their chest was opened with a rib-scissor and the left ventricle was pierced with a 14 G branule. While the inferior caval vein was opened for exsan- guination, the animals were perfused with 300 ml cooled saline. This method allows the heart to perfuse the saline through the circulatory system and makes the perfusion easier. The perfusion of the carotid arteries is an important step that removes blood and potential clots from the vessels as blood cells remaining in the vessels could lead to unreliable molec- ular biolgical results.

From a median cervical section both right and left common carotid arteries were explored, prepared under operational microscope and were excised in full length. The approximate- ly 15 mm long pieces were cleared from the connective and fatty tissue. Two third length of the vessels from every animal was locked in tissue processing cassettes and fixed in 4 % formaledehyde for histology, while the rest was quick-frozen in liquid nitrogen and stored in −80 °C for further quantitative real-time PCR processing.

Histological Analysis

Carotid samples were embedded in paraffin. Morphometric analysis was performed in three hematoxylin-eosin stained cross-sections of each carotid artery by using computer- aided planimetry. Neointima/media ratio and percent of ste- nosis were calculated and compared among groups. For de- tailed description see the online supplementary data.

Immunohistochemical Staining

Immunohistochemical analysis was performed in three repre- sentative cross-sections of each carotid and an average was calculated for each animal. Alpha smooth muscle actin (aSMA), MMP-9 stainings were performed and were evalu- ated by a semiquantitative scoring system by analysts blinded to the study groups. Values represent total scores (area score x intensity score). For a detailed description see the online supplementary data.

Sirius Red Staining

Sirius red staining was performed for the detection of collagen in the media of the uninjured and injured arteries. For a detailed description see the online supplementary data.

Quantitative Real-time Polymerase Chain Reaction (PCR)

Total RNA was isolated from the carotid segments with RNeasy Fibrous Tissue Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA concentration and purity were determined at 260, 280, and 230 nm wave-length with a spectrophotometer. Reverse transcription was performed with the QuantiTect Reverse Transcription Kit (Qiagen, Hilden, Germany) using 400 μg RNA in a volume of 20 μL. Quanti- tative real-time PCR was performed on the LightCycler480 system with the LightCycler480 Probes Master and Universal Probe Library probes (Roche, Mannheim, Germany). The effi- ciency of the PCR reaction was confirmed with standard curve analysis. Every sample was quantified in duplicate, normalized to β-actin expression. Expression of genes involved in inflam- mation, proliferation and extracellular matrix metabolism aSMA, MMP-2, MMP-9, transforming growth factor -β (TGF- β), proliferating cell nuclear antigen (PCNA), protein a20, nuclear factor kappa-b (NF-κB), monocyte chemotactic protein (MCP)-1, Interleukin-6 (Il-6), E-selectin and inducible nitric oxide synthase (iNOS) was determined. For a detailed description see online supplementary data.

Statistical Analysis

Data were tested for normal distribution with Shapiro-Wilk’s test and as they met the requirements for parametric analysis means were tested by one-way analysis of variance (ANOVA) and Tukey’s post hoc test. All data are expressed as mean± standard error of mean (SEM).

Results

Attenuation of Neointima Formation

Zn (ASA)2 significantly lowered neointima formation com- pared to vehicle treatment (stenosis ratio of 25.5±3.9 % in zinc-aspirinate treated rats compared to 54.0±5.8 % in the injured control group) (Fig. 1a). Intimal/medial ratio was also significantly reduced in Zn (ASA)2 treated animals when compared to any other groups (0.5±0.1 vs. injured control: 1.5±0.2; ASA: 1.2±0.2 and Zn: 1.1±0.1 groups) (Fig. 1b).

Expression of Smooth Muscle Alpha–actin

Immunohistochemical staining was performed to determine aSMA expression in medial and neointima smooth muscle cells. Semiquantitative analysis showed a significantly de- creased aSMA presence in the neointimal cells of Zn and Zn (ASA)2 treated groups compared to injured control or ASA (total score of injured control: 6.8±0.3, ASA: 5.6±0.5, Zn:2.8±0.3, Zn (ASA)2: 2.7±0.3). The aSMA presence in medial layers of the vessels was significantly lower in the injured control and ASA groups, whereas there was no remarkable difference among uninjured controls vs Zn and Zn (ASA)2 (relative mRNA expression of uninjured control:7.5±0.4, in- jured control: 2.3±0.2, ASA: 2.5±0.2, Zn: 6.4±0.3 and Zn (ASA)2: 7.4 ± 0.4). aSMA mRNA expression in the Zn (ASA)2 treated group was similar to absolute control values, and significantly higher than in the injured control, ASA or Zn groups (Fig. 2a).

Fig 1 Effects of Zn-aspirinate on injury-induced vascular stenosis. Mor- phometric results and representative cross-sections (hematoxylin-eosin) 18 days after carotid balloon-injury. a) Stenosis ratio (%); b) Neointima/ media area ratio; c) Uninjured control; d) Injured control; e) ASA; f) Zn; g) Zn(ASA)2 (n=12 each). Values represent mean±SEM; *P<.05 vs. uninjured control; #P<.05 vs. injured control.

Down Regulation of MMPs by Zn–aspirinate

Semiquantitative analysis showed significantly lower MMP-9 presence in the medial and intimal layers of the carotides after Zn and Zn (ASA)2 treatment compared to injured control and ASA treated arteries (Fig. 2B). Corresponding with the chang- es in MMP-9 expression, collagen content in the media of the injured control group tended to be lower compared to unin- jured arteries, and higher in the treated carotids (Fig. 2c). Regarding both MMP-2 and MMP-9 mRNA expression, an approximately three-fold up regulation was observed in the injured control group, whereas the mRNA levels in the Zn and Zn (ASA)2 treated carotid arteries were close to normal values (Fig. 3b, 3c).

mRNA Expression of Regulatory Factors TGF-β, PCNA, A20, NF–κB and Adhesion Factor MCP-1

Zn (ASA)2 treatment led to significant down-regulation of TGF-β compared to any other group (Fig. 3D). Eighteen days after injury up-regulated PCNA expression was only detect- able in the injured control group (Fig. 3e). Zn (ASA)2 and Zn treatment significantly up-regulated the mRNA expression of protein A20 and down-regulated the mRNA expression of NF-κB compared to uninjured, injured and ASA treated groups (Fig. 3f, 3G). Compared to uninjured control, signifi- cantly increased MCP-1 expression was detected in the in- jured control group, whereas Zn (ASA)2 treatment significant- ly decreased its level (Fig. 3h) However, no significant differ- ences were detected in the expressions of Il-6, E–selectin and iNOS (data not shown).

Effect of Treatments on Neointimal Apoptosis

TUNEL assay was performed to test a possible contribution of apoptosis to the decreased neointimal hyperplasia. No relevant difference was found in the number of damaged cells by DNA strand breaks in a 1000 μm2 area unit (injured control: 5.1± 0.5; ASA: 4.5±0.4; Zn: 4.8±0.4; Zn (ASA)2: 4.6±0.4).

Fig 2 Representative photomicrographs and semiquantitative scoring of immunohistochemical stainings for a) alpha smooth muscle actin (aSMA, red staining), b) matrix-metalloproteinase (MMP)-9, brown staining) in the medial and neointimal and c) collagens in the medial layer (n=12 each). Values represent mean±SEM; *P<.05 vs. uninjured control; #P<.05 vs. injured control.

Fig 3 Effect of carotid artery balloon-injury and Zn-aspirinate on the relative mRNA expression of a) alpha smooth muscle (aSMA); b) matrix- metalloproteinase (MMP)-9; c) matrix-metalloproteinase (MMP)-2; d) transforming growth-factor (TGF)- β; e) proliferating cell nuclear antigen (PCNA); f) protein A20 (TNFIP3) g) nuclear factor kappa-B (NF-kB) and h) monocyte chemotactic protein-1 (MCP-1). Uninjured controls were given the arbitrary value of 1 and are not shown as bars. Values represent mean±SEM; *P<.05 vs. uninjured control; #P<.05 vs. injured control.

Discussion

In the present study, we examined the inhibitory effect of Zn– aspirinate on the development of restenosis. Compared to the untreated–group, oral treatment with Zn (ASA)2 led to a significant reduction of stenosis ratio and neointima/media ratio in the injured carotides. Molecular biological results support that along with decreased change in the expression of aSMA, zinc–aspirinate treatment resulted in decreased MMP expression in the VSMCs.

Intimal injury or endothelial denudation allows the expo- sure of the thrombogenic subendothelium [5, 18], thus pro- voking a rapid, yet self–limiting platelet activation, leucocyte infiltration, production of reactive oxygen species, smooth muscle cell proliferation, vascular remodeling and thereby restenotic lesion of the vessel. The formation of neointima was successfully hindered by phosphodiesterase (PDE)–5 inhibition [19] and guanylate-cyclase activation [20] which indicates the role of protein kinase G as an upkeeper of status quo in the intact arterial wall through inhibition of apoptotic and oxidative signalling along with proliferation and migra- tion of VSMCs. Another study tested the trace element zinc injected intraperitoneally and found inhibited neointima for- mation in rat carotid arteries [14]. This tendency was also supported by our results after oral administration of zinc, but did not reach the level of significance (P=0.068), and was surpassed by zinc-aspirinate (Fig. 1). The deprivation of zinc led to increased oxidative stress and mitogenic signaling through downregulation of Cu/Zn superoxide dismutase (SOD) [21]. Cell culture experiments revealed that the MAPK-JNK/12 pathway was responsible for the VSMC spe- cific anti-proliferative effect of zinc where zinc supplementa- tion inhibited the induction of NF-κB through the direct induction of zinc-finger protein A20 (also known as tumor necrosis factor-alpha induced protein 3) [21, 22]. These find- ings are supported by our results, where both zinc and zinc- aspirinate treatment led to increased A20 and decreased NF-κB mRNA expression (Fig. 3f, 3G).

In our experiments orally administered acetylsalicylic-acid in the dosage of 90 mg/kg resulted only in a slight decrease of stenosis rate without reaching the level of significance (P= 0.08) (Fig. 1). The low effectivity of acetylsalicylic-acid may be due to its lower affinity to inflammatory regulator cyclo- oxygenase (COX)-2 [23], while through the inhibition of COX-1 isotype it still reduces the release of mediators from platelets in the lesion site. Zn-aspirinate, on the other hand, significantly inhibited neointima formation in the injured ar- teries (Fig. 1).

Acetylsalicylic-acid has propensity to cause ulcers, and Zn2+ is known to possess anti-ulcer as well as anti- inflammatory activities, and a physical mixture of aspirin and zinc did not fulfill the expectation to avoid this side- effect [15]. The advantages of Zn-aspirinate over the physical mixture of aspirin and zinc may be related to its complex form: the free carboxyl group of aspirin is masked. The ulcerogenicity of aspirin may be further reduced by direct gastroprotective action of zinc, which is present in a better tolerated form. Besides, gastrointestinal absorption of aspirin as well as of zinc may be enhanced by the complex form, and zinc may add to the anti-inflammatory effects of aspirine [16]. Aspirine moderates the release of early proinflammatory me- diators from platelets and macrophages (PDGF, TxA2, ADP, COX2 products), thereby hindering the adherence and the migration of inflammatory cells to the intima. These factors also provoke the transformation of VSMCs from their con- tractile phenotype (high aSMA, low MMP and PCNA pres- ence) in the intima and media into a proliferative phenotype (low aSMA, high MMP and PCNA) with high mitogenic, secretory and migratory activity (Fig. 2).By suppressing the release of the inflammatory mediators, aspirin hinders the early transformation of VSMCs. The zinc component of the complex is responsible for the downregulation of interleukines (interleukin (Il)-1β, Il-6, Il-8), enzymes (COX2, inducible nitric oxide synthase), adhesion molecules (E-selectin, intercellular adhesion molecule-1, vascular cell adhesion molecule-1) and growth factors (cycline A, cycline D1) through the inhibition of NF-κB by A20 upregulation and through ROS elimination by SOD and PDE-1 inhibition [13, 22, 24]. Thus, the leveé of inflammation in vessels treated with zinc-aspirinate is less intensive and declines faster. In line with these findings, Zn and Zn (ASA)2 treatment led to significant up-regulation of the zinc-finger protein A20 and to the down-regulation of NF-κB and MCP-1 (Fig. 3F, G, H). Though the peak of inflammation and VSMC transformation/proliferation can be observed 2–7 days after the injury [25], (unlike Il-6, E-selectin and iNOS), MMP-2 and MMP-9 expression remained elevated 18 days after the injury in the untreated and aspirin treated samples, while their levels after zinc-complex treatment were found similar to those of the control group without carotid injury at this time (Fig. 2B-3).

Patients with diabetes mellitus type 2 have a 2 to 4 time’s higher risk of developing coronary artery disease, and have a higher risk of restenosis after successful PTCA. The experi- mental model proved the beneficial effect of zinc-aspirinate in hyperglycemia and metabolic syndrome-like disorders [16]. To the best of our knowledge, we described for the first time the advantageous effect of bis (aspirinato) zinc (II) complex in the prevention of neointima formation in a rat model of endothelial injury. We found that bis (aspirinato) zinc (II) complex combines the anti-proliferative and anti- inflammatory effects of zinc as well as the anti-inflammatory and anti-thrombotic of aspirinate. Due to the synergistic nature of the components in the complex, it could be a promising candidate for diabetic patients with a high risk of restenosis following PTCA.

Study Limitation

The complex interplay among cytokines, growth factors and cells in the vessel wall is underestimated and oversimplified in many animal models, which typically involve a single injury in an artery without preexisting lesions. Moreover, confound- ing metabolic variables, such as diabetes, hypercholesterol- emia or hypertension are often absent in these animals unlike in patients.