| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2008;7:959-962. doi:10.1510/icvts.2008.180489 © 2008 European Association of Cardio-Thoracic Surgery
Carbon monoxide induces relaxation of human internal thoracic and radial arterial grafts
a Department of Cardiothoracic and Vascular Surgery, European Hospital Georges Pompidou, 20 Rue Leblanc, 75015 Paris, France Received 25 March 2008; received in revised form 16 July 2008; accepted 21 July 2008
Corresponding author. Tel.: +33 713 922 2393; fax: +33 713 500 0647.
Carbon monoxide is produced by the degradation of heme by intracellular heme-oxygenase. The aim of our study was to evaluate, in vitro, the vasodilating effect of carbon monoxide and its mechanisms of action on human internal thoracic and radial artery grafts. Segments of human internal thoracic artery and radial artery, obtained from isolated coronary artery bypass surgery patients, were studied in organ chambers. The arterial rings were precontracted with norepinephrine then submitted to carbon monoxide. Inhibitors of nitric oxide synthase and of soluble guanylate cyclase were added to some arterial rings. Carbon monoxide induced significant relaxation in precontracted human internal thoracic artery and radial artery rings. This relaxation was independent of the presence of functional endothelium in internal thoracic artery. Blocking soluble guanylate cyclase partially inhibited this relaxation, while blocking nitric oxide synthase had no effect. Carbon monoxide has a relaxing effect on human internal thoracic artery and radial artery grafts in vitro, partially via cyclic guanylate monophosphate (cGMP) pathway activation. Inducing carbon monoxide production at the cellular level in vivo in human arterial grafts might help prevent vasospasm.
Key Words: CABG; Nitric oxide; Arterial grafts
The current trend in coronary artery bypasss (CAB) surgery is total arterial revascularization, with the internal thoracic artery (ITA) being the graft of choice for all patients. Generalization of the use of arterial grafts such as the radial artery (RA) is still limited by the concern for vasospasticity. Post-operative spasm has been reported to be as high as 5–10% for radial artery [1]. A better understanding of vasoactive properties of human arterial grafts would certainly help prevent spasm and improve permeability in the clinical setting of CAB surgery. Heme-oxygenase (HO) plays an important role in the regulation of arterial tone as well as in cellular protection. We have shown in a previous study that the induction of HO reduces contractility in human arterial grafts [2]. Carbon monoxide (CO) seems to mediate the vasoactive properties of HO. Some animal reports have shown that exogenously administered CO relaxes isolated blood vessels [3–5]. This effect is not unanimously reported [3–5]. The direct effect of CO on human arterial grafts has never been studied. The aim of our study was to evaluate the vasoactive role of CO on human ITA RA grafts in vitro, and to define the mechanism of action of CO.
The study was in accordance with the principles outlined in the Declaration of Helsinki and was approved by the local Ethical Committee. All patients gave their informed consent. Table 1 shows the clinical characteristics of these patients.
2.1. Sampling of the arterial segments Human ITA (10 segments) and RA (five segments) were obtained intraoperatively from 11 patients undergoing elective isolated CAB surgery. The harvesting technique for both ITA and RA was described in an earlier report [2]. Immediately after harvesting, the arterial segments were immersed in a Dulbecco-modified Eagle's solution (DMEM, Life Technologies, Scotland) at 4 °C. The mean time between harvesting and experimentation was 45 min. The arterial rings were studied in organ bath chambers. Care was taken not to traumatize the endothelium. The number of rings taken from each blood vessel varied from 4 to 10. The organ bath experimentation technique has been described in previous reports [2, 6]. Arterial rings were mounted on two stainless-steel hooks in a 20 ml water-jacketed glass organ bath filled with physiological Krebs solution maintained at 37 °C (Krebs solution had the following composition (M): NaCl 118; KCl 4.7; CaCl2 2.5; KH2PO4 1.2; MgSO4 1.2; NaHCO3 25; EDTA 0.026; dextrose 11.1). The solution was continuously oxygenated with carbogen gas (95% O2; 5% CO2). In order to study the tension response of the arterial rings independently of their varying diameter, the arterial rings were progressively stretched to their respective resting tension: 5 g for the IMA rings and 9 g for the RA rings according to the work of Chardigny et al. [6]. 2.3. Vasoreactivity experimentation protocol First, the maximal contractility of the arterial rings was evaluated in the presence of KCl 10–2 M. The arterial rings were rinsed afterwards to their basal tension.Second, the functionality of the endothelium was verified by means of the endothelium-dependant relaxing effect of acetylcholine (Ach) 10–6 M and calcium ionophore 10–6 M [7–9]. Following this step, the arterial rings were classified into two groups according to the presence or absence of functional endothelium. Third, the arterial rings were contracted to 60–80% of their maximal contraction using norepinephrine (NE) (3.10–7 M). Once the contraction stabilized, the carbogen infusion was stopped and the organ chambers were bubbled with CO (100%) at a rate of 2 ml/min, until tension stabilization. At this flow of CO, the maximal concentration of CO in the organ bath would have reached 3.10–5 M according to the work of Hussain et al. [10]. To make sure the relaxing effect of CO was not due to the transient hypoxia secondary to the discontinuation of carbogen, the control rings were bubbled at the same rate with a low oxygen gas mixture (LOGM) (5% O2; 95% CO2). By submitting both CO-treated and control rings to the same condition of hypoxia, the difference in relaxation between the two groups would only be due to CO treatment. Fourth, after rinsing and stabilization, some arterial rings were incubated for 30 min in the presence of a selective inhibitor of nitric oxide synthase (NOS) [Nw-Nitro-L-Arginine (LNA) at 10–4 M], or a selective inhibitor of soluble guanylate cyclase [1H-(1,2,4)oxadiazolo[4,3,-a]quinoxalin-1-one (ODQ) at 10–5 M]. The complete blockade of NOS with LNA was verified by the absence of Ach-induced relaxation and the persistence of nitroglycerin-induced relaxation (NTG 10–6 M, which is a donor of NO). The complete blockade of soluble guanylate cyclase (sGC) with ODQ was verified by the absence of relaxation in response to NTG 10–6 M. Control arterial rings were incubated in the same conditions without the addition of inhibitors. Finally, after incubation, the arterial rings were again contracted to 60–80% of their maximal contraction using norepinephrin (3.10–7 M). The organ chambers were bubbled with either CO or LOGM until stabilization. After discontinuation of CO and reestablishment of carbogen, a maximal relaxation was obtained using papaverine (10–4 M). 2.4. Data and statistical analysis Tension generated by arterial rings was digitalized by a computer using the IOX software (EMKA Technologies, France). The results were expressed as percentage of the maximal initial contraction induced by NE. Maximal relaxation induced by papaverin was used as reference.In each experimentation, CO-treated rings and the control rings were harvested from the same patient. For all studied parameters, n represented the number of patients. Results were expressed as mean±standard error of the mean. Statistical analysis was performed with Student's t-test for paired values or two-way analysis of variance (ANOVA) followed by Bonferroni test for evaluating the effect of different treatments.
3.1. Effect of CO on human ITA In the presence of functional endothelium, CO caused a significant and sustained relaxation of the human ITA rings (Fig. 1a). This relaxation started 1 min after CO exposure, and reached a maximum at 6 min. The ITA rings rapidly recovered their initial contraction tension after discontinuation of CO and reestablishment of carbogen. This CO-induced relaxation was reproducible with time (the relaxation was statistically comparable during the first part and second part of the experimentation on two-way analysis of). The control rings exposed to LOGM presented a slow and limited relaxation compared to the CO-treated rings. The absence of functional endothelium did not change the CO-induced relaxation.
The addition of LNA did not change the effect of CO (Fig. 1b). The addition of ODQ completely inhibited the relaxation induced by CO in the first few minutes of exposure, then after 5 min the inhibition became incomplete (Fig. 1c). All the studied RA rings had functional endothelium verified by the Ach-induced relaxation test. Similarly to ITA, CO caused a significant and sustained relaxation of the human RA rings compared to control (Fig. 2a). The RA rings rapidly recovered their initial contraction tension after discontinuation of CO. The CO-induced relaxation was reproducible with time.
The addition of LNA did not change the effect of CO (Fig. 2b). The addition of ODQ inhibited partially the relaxation induced by CO. This inhibition became significant after 5 min of exposure (Fig. 2c). 3.3. Comparison between human IMA and RA The tension curves of the control ITA and RA rings were statistically identical. CO induced a more important relaxation of human RA compared to ITA (Fig. 3).
CO induced an endothelium-independent relaxation in precontracted human ITA and RA rings. This relaxation was not mediated by the production of NO, since it was not modified by the addition of LNA. This reinforces the fact that, like NO, CO has a vasorelaxing effect on human arteries. To verify whether the vascular effect of CO was mediated by cyclic Guanylate monophosphate (cGMP) production, we completely blocked the sGC using ODQ. Our study showed that ODQ had variable effects on CO-induced relaxation, depending on the nature of the studied arteries. In human ITA rings, ODQ blocked the initial relaxing effect of CO. After a few minutes, the persistent stimulation with CO and/or the augmentation of CO concentration in the organ bath succeeded in inducing partial relaxation of human ITA. This relaxation was most likely due to the stimulation of alternative pathways, other than the cGMP pathway. In human RA rings, ODQ inhibited partially the relaxation induced by CO. This inhibition became significant after 5 min of exposure, proving again the existence of both cGMP-dependant and cGMP-independent pathways. This observation cannot be explained by an incomplete inhibition of sGC by ODQ, since as we already stated, ODQ efficacy was verified prior to CO administration. In an animal study, Wang et al. [11] reported similar endothelium-independent CO-induced vasodilation in rat tail arteries. In his study, a complete inhibition of the effect of CO could only be achieved by the association of a c-GMP inhibitor and an inhibitor of Ca-dependent K channels (such as charybdotoxin). Another study by Naik and Walker [12] also reported a cGMP-independent action by CO. The conclusions of these two studies reinforce ours and help explain the incomplete inhibition of CO-induced relaxation by ODQ. In contrast, other animal studies [10, 13–15] showed a vasorelaxing effect of CO which was completely reversible by inhibitors of c-GMP pathway. All these previous reports were on animal arteries. Our study showed for the first time that exogenous CO had a relaxing effect on human ITA and RA. This explains our earlier results [2] showing that the induction of HO decreases, via CO production, the contractile force in human ITA and RA. The present study also suggested the existence of both c-GMP-dependent and c-GMP-independent pathways to explain the effect of CO. The preponderance of one or the other pathway depends probably on the type of artery and of species studied. The relaxing effect of CO is not unanimously reported. Brian et al. [3] did not find any relaxing effect of CO (at concentration lower than 10–4 M) on rabbit and dog cerebral arteries. Only a moderate relaxation was observed at higher concentration (3.10–4 M). Andresen et al. [4] did not observe any relaxation in rat middle cerebral arteries exposed to CO. In the same experiment, precontracted rat gracilis arteries responded differently. Another animal study by Johnson and Johnson [5] reported that CO at a concentration of 5.10–5 M induced vasoconstriction by interfering with NO synthesis. Carbon monoxide induced a more important relaxation in human RA compared to ITA. CO could play a more prominent role in vivo in the control of vasoreactivity in RA compared to ITA. The concentration of CO in organ bath was not measured in this study, but was already determined by Hussain et al. [10] who studied the concentration of CO according to the rate of CO bubbling in organ chamber. CO concentration had no direct implication on the interpretation of our results, since the aim of our study was to verify that CO had a relaxing effect on human IMA and RA as a first step, and to compare this relaxing effect in the presence of different inhibitors. We are currently conducting studies using CO releasing molecules (Tricarbonyldichloro Ruthenium (II) dimer [Ru(CO)3Cl2]2 or CORM-2) at increasing concentrations; that would allow us to establish a dose–response relaxation curve in the presence of CO. In conclusion, our study confirmed that exogenous CO induces relaxation in human ITA and RA, in vitro. The RA was more sensible to CO than ITA. This action was independent from the presence of endothelium, and from the NO pathway. CO vasorelaxation is partially induced by stimulating c-GMP formation. Other mechanisms of action are also responsible for the effect of CO, and these pathways may be differentially activated depending on the nature of the vessel studied. Further studies, using CORM-2 are being conducted to better understand the mechanisms of action of CO, and its effect on the production of cGMP. Molecules that might potentiate the relaxing effect of CO on human arteries are currently being investigated. Inducing local CO production and/or potentiate the effect of CO in human arterial graft might help prevent spasm.
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
