YC-1, a Benzyl Indazole Derivative, Stimulates Vascular cGMP and Inhibits Neointima Formation

Abstract

The pathobiologic process of arterial stenosis following balloon angioplasty continues to be an enigmatic problem in clinical settings. This research project investigates the ability of YC-1, a benzyl indazole derivative that sensitizes sGC/cGMP, to stimulate endogenous cGMP and attenuate balloon injury-induced neointima (NI) formation in the rat carotid artery. Northern and Western blot analyses revealed enhanced acute expression of iNOS and inducible heme oxygenase (HO-1) mRNA and protein in the injured artery. The contralateral uninjured artery also demonstrated acute HO-1 mRNA and protein induction without detectable iNOS expression. Perivascular application of YC-1 immediately following injury significantly stimulated acute vessel wall cGMP compared to untreated controls. YC-1 treated sections demonstrated significant reduction in NI area (−74%), NI area/medial wall area (−72%), and NI thickness (−76%) 2 weeks post-injury. These results directly implicate YC-1 as a potent new therapeutic agent capable of reducing post-angioplasty stenosis through endogenous CO- and/or NO-mediated, cGMP-dependent processes.

The arterial remodeling response to endovascular injury is manifested as neointimal formation with a major diminution in luminal patency. While certain characteristics of this adaptive response to arterial injury have been characterized, the regulation of this pathologic process remains largely unknown and continues to contribute to the clinical problem of post-angioplasty stenosis. Nitric oxide (NO), generated from L-arginine through the action of either constitutive or inducible NO synthase (NOS), is a well established, potent vascular mediator that has been suggested to modulate smooth muscle cell (SMC) phenotype (1) and the arterial response to endovascular injury (2) through cGMP-dependent processes. More recently, carbon monoxide (CO), another diatomic gas, has been shown to serve physiological roles in platelet function (3), vasomotor tone regulation (4), and in the control of inflammation (5) and SMC proliferation (6) under inimical conditions. CO is liberated as a byproduct of the enzymatic action of either inducible (HO-1) or constitutive (HO-2 or HO-3) heme oxygenase on heme that yields free iron and biliverdin. Similar to NO, CO activates soluble guanylyl cyclase (sGC) to stimulate cGMP production (7), exerting its actions in both autocrine and paracrine fashion. However, the potency of CO in stimulating sGC and cGMP production is markedly lower than that of NO. CO stimulation leads to a 4- to 6-fold activation of the purified enzyme, while NO stimulation of the enzyme reaches 400-fold (8). Despite this relative lack of potency, CO potentially exerts a myriad of vascular cell functions, many of which mimic those of NO (reviewed in 9). Different responses to regulatory inhibitors and inducers of NOS and HO, as well as significant biochemical differences between the two gases, however, reveal that these systems do indeed represent separate physiological signaling mechanisms (7, 9).

A potentially physiologically relevant mechanism that sensitizes sGC and cGMP has recently been described, and may represent an important new therapeutic intervention. YC-1 [3-(5′-hydroxymethyl-2′-furyl)-1-benzyl indazole] is a chemically synthesized compound that was originally shown to directly stimulate platelet sGC activity and cGMP production and thereby inhibit platelet aggregation and thrombosis (10). YC-1 also stimulates vascular endothelial and SMC sGC and cGMP, independent of the actions of NO (11, 12). Furthermore, YC-1 potentiates NO- and CO-induced sGC activation (13) and enhances endothelial NO synthesis and release (14). It has been suggested that the mechanism of action of YC-1 is to stabilize the active configuration of sGC (11) and to decrease the dissociation rates of the gases from the activated enzyme (8). Through these mechanisms, YC-1 is capable of stimulating CO-induced sGC with a potency and specific activity similar to that attained with NO stimulation (8, 15). In addition, YC-1-mediated inhibition of the cGMP-specific phosphodiesterase type 5 (PDE-5) has been shown to occur in platelets (13) and in aortic extracts (16). This ability of YC-1 to directly activate sGC, to potentiate the stimulatory actions of NO and CO on sGC, and to concurrently cause a persistent elevation of cGMP (through inhibition of cGMP breakdown) makes it a potent activator of the sGC/cGMP system and a potentially attractive new therapeutic agent. The existence of an endogenous “YC-1-like substance” has been suggested that would be capable of synergizing with endogenous CO and/or NO to stimulate cGMP under pathophysiologic conditions.

The purpose of this study was to investigate the ability of YC-1 to (a) stimulate endogenous vascular cGMP synthesis and (b) attenuate the neointimal response to endovascular injury. We report induction of local iNOS and local and systemic HO-1 following balloon injury of rat carotid arteries. Significant acute upregulation of arterial cGMP from YC-1 treatment was observed, and this was associated with a subsequent and robust diminution in injury-induced neointima formation. We hypothesize that endogenous cGMP may represent an important in vivo regulator of cardiovascular function and response, and that YC-1 may provide a potential new therapeutic strategy in interventional angioplasty.

METHODS

Rat carotid artery balloon angioplasty.

We utilized the established rat carotid artery (CA) model of balloon angioplasty to examine the in vivo arterial response to injury (17). Male Sprague Dawley rats (Harlan, Indianapolis, IN) weighing an average 497 g were anesthetized with a combination anesthetic (ketamine, zylazine, and acepromazine; 0.5–0.7 ml/kg, IM; VetMed Drugs, Houston, TX), and the left CA and external carotid branch exposed. A Fogarty 2F embolectomy catheter (Baxter Healthcare Corp., Irvine, CA) was introduced into the external carotid branch through an arteriotomy incision and advanced to the aortic arch. The balloon was inflated and withdrawn three times with rotation. The catheter was removed and the external carotid ligated. The overlying tissue was sutured and the skin closed with rodent wound clips. After full recovery from anesthesia, animals were returned to the animal care facility and provided standard rat chow and water ad libitum. Rats were sacrificed by pneumothorax and exsanguination, and the tissues harvested for use in specific protocols.

YC-1 dosing.

Immediately following balloon injury, 200 μl of a 25% solution of a copolymer gel (Pluronic F-127; BASF Corporation, Mount Olive, NJ) plus YC-1 (1 mg in 50μl DMSO; Calbiochem, La Jolla, CA) was administered topically to the exposed adventitia of the distal CA section. Control animals received 200 μl empty gel with 50 μl DMSO on this distal CA section. The scheme for YC-1 administration is shown in Fig. 1. In all experiments, the entire length of the left CA was injured (avg. 22 mm), while the contralateral right CA served as an unmanipulated untreated control. Only the distal half of the injured left CA was exposed during the surgery, and this section (approximately 11 mm) was treated with YC-1 (or empty gel) while the proximal half remained untreated. A cohort of animals that was balloon-injured and not exposed to any gel or other treatment served as a separate control group.

FIG. 1.

Scheme for YC-1 administration on balloon injured left carotid artery (CA). The entire length of the left CA was balloon injured, immediately followed by perivascular application of YC-1 (1 mg) on the distal section. Sites where tissues were taken for “distal + YC-1” and “proximal − YC-1” are indicated. The right CA served as an uninjured control. EC, external carotid branch; IC, internal carotid branch.

Tissue processing and staining.

Acute studies examining enzyme and cGMP induction used freshly obtained tissues that were removed from the sacrificed animal and immediately snap-frozen. For the two week studies examining morphologic remodeling of the vessel wall, animals were perfusion-fixed transcardially using warmed PBS followed by 10% buffered formalin phosphate. Tissues were fixed, processed by standard procedures, and paraffin-embedded. Five micron sections were cut using a rotary microtome and placed on pretreated slides. For Verhoff’s elastic tissue staining, slides were deparaffinized and stained with Verhoff’s solution of alcoholic hematoxylin, ferric chloride, and potassium iodine. Slides differentiated with ferric chloride and counterstained with Van Gieson’s solution containing 1% acid fuchsin and picric acid.

Morphologic analysis.

Microscopic quantitation of vessel dimensions was performed using Zeiss Image 3.0 (Media Cybernetics) and Adobe Photoshop 4.0 (Adobe Systems) software systems linked through a CCD color camera (Leaf Microlumina, Leaf Systems, USA) to a Zeiss Axioskop 50 light microscope (Carl Zeiss, Germany). Images were measured for perimeters and areas corresponding to internal and external elastic laminae and lumen. Numerical transformations provided data for neointimal and medial wall areas and vessel diameters.

cGMP RIA.

Cyclic GMP levels were measured in vessel samples with a competitive RIA using ¹²⁵I-labeled cGMP, following instructions from the manufacturer (Amersham Pharmacia Biotech, Piscataway, NJ). Tissues were freshly removed and immediately snap-frozen. Tissues were homogenized in cold 6% trichloroacetic acid, centrifuged, and the cGMP-containing supernatant removed and ether-washed. The cGMP extract was dried and resuspended. Samples were acetylated, and working standards prepared using serial dilutions for RIA. Separate sections of thoracic aorta incubated in 10 μm sodium nitroprusside, a potent sGC activator, for 4 min at 37°C provided a positive control for cGMP (18).

Northern blot analyses.

Total tissue RNA was extracted from arteries with guanidium isothiocyanate, and RNA (10 μg) was fractionated on 1% formaldehyde agarose gels and transferred to nitrocellulose membranes. The filters were hybridized with cDNA probes specific for rat HO-1 and iNOS (19, 20). The cDNA fragments were labeled with [α-³²P]dCTP using a standard random-primed reaction. Membranes were then hybridized with the cDNA probes as previously described (21), washed twice, and then exposed to Kodak X-Omat film. The membranes were subsequently stripped and rehybridized with a [³²P]-labeled GAPDH probe.

Western blot analyses.

Arterial tissues were homogenized in cold lysis buffer (125 mmole/L Tris–HCl [pH 6.8], 2.5% DTT, 2% SDS, and trace bromophenol blue), boiled, sonicated, and SDS–PAGE was performed on 10% gels for iNOS and 20% gels for HO-1. The blots were electrophoretically transferred to nitrocellulose membranes and blocked for 1 h. Membranes were incubated with either a polyclonal antibody specific for macrophage iNOS (1:2000; Transduction Laboratories, Lexington, KY) or HO-1 (1:500, Stressgen Biotechnologies Corp., Canada) in PBS for 1 h at room temperature. After incubation with appropriate secondary antibodies, blots were incubated in enhanced chemiluminescence reagents (Amersham Corp.) and exposed to photographic film.

Statistical analyses.

All data are represented as mean ± standard error (SE) of the mean. A two-tailed paired Student’s t-test was used to compare same animal data, while all other data were grouped according to treatment and analyzed using an unpaired Student’s t-test. P-values < 0.05 were considered statistically significant for all comparisons.

RESULTS

As shown in Fig. 2, Northern blot analyses demonstrate upregulation of both iNOS and HO-1 mRNA in the ipsilateral left CA (L) 24 h following balloon injury. The contralateral right CA (R) also demonstrates induced HO-1 mRNA at this time point; however, no detectable iNOS signal is evident in the right CA (R). Western blot analyses, shown in Fig. 3, similarly reveal induced iNOS and HO-1 protein in the injured left CA (L) 24 h post-injury. The right CA (R) also shows elevated HO-1 protein and no detectable iNOS protein expression. CA from control (C) rats not subjected to balloon injury do not express iNOS or HO-1 mRNA or protein.

FIG. 2.

Northern blot analysis of iNOS and HO-1 mRNA expression 24 h following left CA balloon injury. Lanes represent the ipsilateral injured left CA (L), the contralateral uninjured right CA (R), and CA from control (C) rats which were not subjected to balloon injury. GAPDH is included as a loading control. Between 8 and 10 animals were used per lane to obtain sufficient amounts of RNA.

FIG. 3.

Western blot analysis of iNOS and HO-1 protein expression 24 h following left CA (L) balloon injury. Lanes are labeled as in Fig. 2.

The injured CA distal sections exposed to YC-1 exhibit a significant (P = 0.012) increase in cGMP content compared to proximal sections (not exposed to YC-1) from the same vessel (1.23 ± 0.13 vs 0.64 ± 0.11; Fig. 4). The CA sections from a separate group of uninjured rats do not show YC-1-induced alterations in cGMP (data not shown). Sections of thoracic aorta were exposed ex vivo to 10 μm sodium nitroprusside, an activator of sGC, for 4 min at 37°C. Sodium nitroprusside stimulated approximately a 2-fold increase in vessel wall cGMP levels (data not shown), similar to that induced by YC-1 in the injured CAs in vivo.

FIG. 4.

Cyclic GMP levels measured by RIA in balloon injured left CA sections 24 h post-injury. Distal arterial sections exposed to YC-1 immediately following injury express a significant (P = 0.012) increase in cGMP content compared to the proximal sections from the same vessel not exposed to YC-1. Values represent mean ± SE. n = 5 animals. *Statistically significant with P < 0.05.

Representative photomicrographs of balloon injured perfusion-fixed left CAs two weeks post-injury are shown in Fig. 5. Figure 5a shows a Verhoff’s elastic tissue-stained cross-section of a balloon-injured left CA. This injured proximal arterial section was not exposed to gel or any other pharmacologic treatment. A significant and concentric neointima is evident, and the elastin-rich laminae are stained black. Collagen is stained dark red, and is highly expressed in the abundant adventitia. Figure 5b illustrates an elastin-stained distal cross-section from the same balloon-injured left CA as shown in Fig. 5a; however, this distal section was exposed to YC-1 (1 mg) in gel immediately after surgery. A significantly attenuated neointima is evident, appearing sporadically as a thin layer adjacent to the black internal elastic laminae. Also evident is a decreased adventitia and reduced collagen staining. Figure 5c illustrates a distal section of a left CA two weeks after injury which had been exposed to empty gel (without YC-1) at time of surgery.

FIG. 5.

Representative photomicrographs of balloon-injured, perfusion-fixed, elastin-stained left CA cross-sections 2 weeks post-injury. (a) shows an injured proximal left CA section not exposed to either gel or any other pharmacologic treatment. A significant and concentric neointima is observed, with elastin fibers of the laminae stained black. Collagen is stained dark red, and is expressed in the abundant adventitia. (b) illustrates an injured distal left CA section from the same vessel as shown in (a), except this section was treated immediately after surgery with 1 mg YC-1. A significantly attenuated neointima is evident, appearing sporadically as a thin layer adjacent to the black internal elastic laminae. Also evident is a decreased adventitia and reduced collagen staining. (c) shows an injured distal left CA section exposed to empty gel (+ DMSO, without YC-1) immediately following surgery. Magnification for all photomicrographs is 100×.

Figure 6 represents morphometric analyses for (a) neointimal cross-sectional area, (b) neointimal area/medial wall area ratio, (c) neointimal thickness, and (d) medial wall area. Data for balloon injured left CA distal sections exposed to YC-1 (“distal +YC-1”) and for balloon injured left CA proximal sections not exposed to YC-1 (“proximal −YC-1”) are shown. The neointimal cross-sectional area exhibits significant attenuation (−74%; P < 0.001) in the distal sections exposed to YC-1 compared to the proximal unexposed sections from the same vessel. Similarly, both the neointimal area/medial wall area ratio (−72%; P = 0.001) and the neointimal thickness (−76%; P < 0.001) show significant attenuation in the distal YC-1 treated sections compared to the proximal unexposed sections. The medial wall cross-sectional area also shows significant diminution (−12%; P < 0.05) in the distal sections exposed to YC-1 compared to the proximal unexposed sections from the same vessel. No major differences were detected between these sections in the control animals for any of these parameters. No differences were detected in the lengths of internal elastic laminae and external elastic laminae between and among all treatment groups (data not shown).

FIG. 6.

Morphometric analyses for (a) neointimal cross-sectional area, (b) neointimal area/medial wall area ratio, (c) neointimal thickness, and (d) medial wall area. Data for injured left CA distal sections exposed to YC-1 (“distal + YC-1”) and injured left CA proximal sections not exposed to YC-1 (“proximal − YC-1”) are illustrated. Values represent mean ± SE. For all parameters n = 6 animals in the control group and n = 9 animals in the YC-1 treated group. *Statistically significant with P < 0.05. ***Statistically significant with P < 0.001.

DISCUSSION

The results of this study implicate endogenous cGMP as an important physiologic regulator of the arterial response to injury. Balloon injury was shown to induce both iNOS and HO-1 mRNA transcript and protein expression in the instrumented CA within 24 h. Perivascular treatment of the injured CA with YC-1 immediately following surgery significantly increased arterial cGMP levels 24 h later. This was associated with a subsequent and robust diminution in injury-induced neointima and vessel wall area after 2 weeks. These results strongly suggest a novel and potentially important therapeutic role for YC-1 in potentiating CO- and/or NO-induced cGMP processes involved in the neointimal response to endovascular injury.

Cyclic GMP is an important intracellular second messenger that is involved in a variety of functional processes. NO binds to the heme moiety of sGC, activating the enzyme to convert cytosolic guanosine triphosphate (GTP) to cGMP. Cyclic GMP mediates its intracellular effects by activation of specific cGMP-dependent protein kinases (PKG). In this study we have shown that local vascular iNOS is upregulated at both the mRNA and protein levels within 24 h following balloon injury. This is consistent with previously published reports of medial wall iNOS mRNA and protein upregulation following carotid artery balloon injury (22, 23). In addition to NO, CO provides an alternate physiologic pathway to stimulate cGMP synthesis. The primary source of endogenous CO production is HO-mediated heme degradation, which produces biliverdin and free iron and liberates CO. The marked upregulation of HO-1 by balloon injury indicates that endogenous vascular CO production is highly stimulated following intraluminal injury. Like NO, CO activates sGC to stimulate cGMP production and initiate subsequent cellular functions (7, 9).

The effects of NO in modulating the arterial response to injury have recently been examined. In rat ileofemoral artery injury, the extent of neointimal thickening was attenuated with application of a NO-releasing agent (2). In another study, eNOS gene transfer resulted in reduction of injury-induced SMC proliferation and neointimal formation in rat CA (24). We and others (17, 25) have recently demonstrated that systemic administration of an HO-1 inducer attenuates the neointimal response to arterial balloon injury. Togane et al. (26) have recently examined the influence of endogenous CO on this adaptive response using in vitro differential gas trapping experiments. These authors suggested CO as the mediator responsible for these protective phenomena. Our current study extends these observations and provides the first report using direct YC-1-induced manipulation of endogenous cGMP to modify injury-induced vascular remodeling.

The rationale for the YC-1 experiments was based on our initial observations that both major sources for endogenous cGMP, NOS-mediated NO and HO-mediated CO, are induced by arterial injury (Figs. 2 and 3). YC-1 is a newly synthesized compound that sensitizes the sGC/cGMP system by stabilizing the active configuration of sGC (11) and by decreasing the dissociation rates of diatomic gases from the activated enzyme (8). YC-1 also has the capacity to inhibit PDE-5 in vascular cells, thereby decreasing cGMP breakdown (13, 16). In this study, perivascular application of YC-1 on the injured CA significantly stimulated vessel wall cGMP synthesis. YC-1 stimulated cGMP only in vessels exposed to balloon injury, presumably because of increased release of both iNOS-mediated NO and HO-1-mediated CO. Associated with this cGMP induction, injured arterial sections exposed to YC-1 demonstrated striking diminution in the extent of arterial remodeling after 2 weeks. These protective phenomena were manifested primarily as reduced neointima and diminished luminal stenosis of the injured artery.

Several mechanisms for these potent protective actions of YC-1 on arterial remodeling can be suggested based on results from in vitro studies. The sGC/cGMP system is markedly potentiated and sensitized to the actions of both CO and NO by YC-1. Cyclic GMP induces the anti-mitogenic cyclin-dependent kinase (cdk) inhibitor p21 (27), inhibits cyclin D1 expression and cdk4 activation, and delays growth factor-induced cdk2 activation through a transient increase in p27 (28). These actions of cGMP potentially suppress the vascular cell cycle machinery necessary for proliferative neointima formation. Alternatively, YC-1 has been found to prevent collagen-induced mobilization of intracellular Ca²⁺ and actin polymerization (29), thereby decreasing vascular cell migration and possibly inhibiting establishment of a pathologic neointima. Vascular cell apoptosis following balloon angioplasty has been well documented (30, 31) and may provide a necessary regulatory process influencing size and stability of vessel wall lesions (30). In addition, delayed upregulation of anti-apoptotic genes may enhance cell viability in the newly established vascular lesion (32). Since the sGC/cGMP system has been suggested to be involved in NO-induced apoptosis (33), YC-1 may be capable of altering the extent of vascular cell apoptosis following balloon injury. It is likely that several of these complex molecular mechanisms are involved in the striking protective effect of YC-1 on the adaptive response to vascular injury.

In conclusion, the findings of our study offer the potential for a novel therapeutic strategy aimed at decreasing post-angioplasty stenosis in clinical settings. YC-1 represents an attractive therapeutic agent with redundant protective mechanisms that regulate endogenous cGMP metabolism and attenuate post-injury arterial remodeling.