The present study demonstrated that even at a low concentration (6 nM), acute mercury administration alters the endothelial function of conductance vessels, as previously reported for resistance vessels 
. Reduction of NO after exposure to mercury promoted decreased the NO endothelial modulation due to increased O2−
production increasing the vasoconstrictor response to phenylephrine. Administration of SOD and the NADPH oxidase inhibitor, apocynin, reduced this increased vasoconstrictor response. Results from pharmacological interventions with losartan, enalapril, indomethacin, NS 398 and furegrelate suggest that the renin-angiotensin system and COX-2 might be involved in the vascular effects of mercury. In addition, our results show an increase in COX-2 protein expression and a decrease in AT1 protein expression.
The fact that the endothelium is affected by low concentrations of heavy metals below the reference values highlights the importance and the need to better understand the mechanisms by which these metals promote development of cardiovascular diseases 
. Mercury induces toxicological consequences as a result of exposure to mercury vapour 
, the release of mercury from or during removal of amalgam fillings 
and the ingestion of contaminated fish 
. Exposure to mercury vapour and removal of amalgam fillings increase blood and plasma concentrations of the metal (4 to 5 nM), but the toxicological consequences are still a matter of debate. In a recent biomonitoring study in New York City adults, the blood mercury concentration was 2.73 ng/mL, whereas in regular fish consumers it reached 5.65 ng/mL 
. In the present study, we used a low concentration of HgCl2
1.6 ng/mL), which is similar to the levels reached in exposed humans.
In some pathological conditions, such as atherosclerosis and hypertension, a reduction in NO bioavailability is reported to contribute to the appearance of cardiovascular disease 
. Thus, a decrease in NO production could contribute for the development of cardiovascular disease. The main factor that contributes to the reduction of NO bioavailability is the increase of ROS. The superoxide anion interacts with NO and forms peroxynitrite, thereby decreasing the NO bioavailability for smooth muscle relaxation 
. Thus, the reduced NO bioavailability induced by mercury could be related to increased levels of ROS. In fact, previous studies have associated mercury exposure with increased oxidative stress 
NO reduction is reported to increase vascular reactivity to vasoconstrictors 
. We observed that acute administration of HgCl2
promoted an increase in reactivity to phenylephrine in aortic rings. The comparison of dAUC changes of aortic rings with and without endothelium, suggested that the magnitude of the response to phenylephrine was lower after incubation with mercury. This result suggests that the ability of the endothelium to negatively modulate the contractile response induced by phenylephrine seems to be impaired in isolated aortic rings incubated with HgCl2
We then investigated if the reduction of endothelial modulation after incubation with mercury in aortic rings could be a consequence of reduced NO bioavailability as previously reported in the tail vascular bed and in mesenteric and aortic arteries 
. In this study we also found reduced levels of NO after exposure to HgCl2
. Grotto and colleagues 
also reported a depletion of NO in rats sub-chronically exposed to MeHg at a low level.
However, an underlying mechanism associated to those previous findings is the oxidative stress generation. Our results showed that treatment with SOD or apocynin were able to reverse the effects of mercury on the vascular reactivity to phenylephrine in aortic rings, suggesting that acute exposure to mercury is accompanied by an increased production of ROS. Therefore, we suggest that increased vascular reactivity to phenylephrine induced by mercury could have been caused by an increased release of ROS with the resulting reduced bioavailability of NO.
We previously reported that treatment with small doses of mercury increases the release of COX-2 derived vasoconstrictor prostanoids and its participation in vasoconstrictor responses 
. The increased activation of the renin-angiotensin system after mercury treatment would be associated with this increased COX-2 activity. These mechanisms would contribute to explain the endothelial dysfunction and the increased vasoconstrictor responses induced by mercury exposure 
. A previous report has shown that an increase of the activity of the local renin-angiotensin system and production of prostanoid vasoconstrictors could increase ROS in the aorta of normotensive (WKY) and spontaneously hypertensive rats (SHR) 
. Therefore, we investigated whether the vascular actions induced by mercury could involve the renin-angiotensin system and the cyclooxygenase pathway. Our findings indicate that HgCl2
is able to increase the release of angiotensin II, which in turn could promote down regulation of AT1
protein expression. In addition, angiotensin II can regulate COX-2 protein expression, the production of prostanoids 
, the ROS production 
and NADPH oxidase activity 
. Thus, increased local release of angiotensin II induced by mercury could cause the increased activity of COX-2 and NADPH oxidase and this could increase the release of ROS, as observed in this study.
In addition, we observed that indomethacin, NS 398 and furegrelate were able to reduce the effect of mercury on the contractile response to phenylephrine. These results suggest that vasoconstrictor prostanoids, specifically thromboxane A2, participate in the increased reactivity to phenylephrine induced by mercury. Corroborating these results, we also observed that acute exposure to HgCl2
increased the protein expression of COX-2. The increased protein expression of COX-2 may result in an increased release of vasoconstrictor prostanoids and in an overproduction of ROS 
. Taken together, these factors might contribute to inflammation, vascular injury and atherosclerosis development, rendering the vascular hyperreactivity to phenylephrine induced by mercury.
Although our results suggest a reduction of NO bioavailability induced by mercury, relaxations induced by acetylcholine and sodium nitroprusside were not altered after incubation of aortic rings with this metal. These results indicate that the release of NO stimulated by acetylcholine was unaffected by incubation with mercury, suggesting that mercury does not seem to acutely modify the activity of NOS or signalling pathways involving guanylate cyclase. In contrast, a previous study has reported alterations in NOS activity due to sub-chronic exposure to methyl mercury 
. Kishimoto and colleagues 
also reported inhibition of NOS in cultured human umbilical vascular endothelial cells as a result of mercury toxicity. However, these studies were performed after sub-chronic exposure to mercury 
or exposure to high concentrations of this metal 
However, our findings do not allow us to define which mechanism is the main one induced by mercury to increase oxidative stress, at least under the present experimental conditions. NO bioavailability reduction caused by ROS or increased endothelial production of vasoconstrictors, angiotensin II and tromboxane A2, altogether increase vascular reactivity.
In summary, these results show that in vitro exposure to low concentrations of HgCl2 (6 nM) increased vascular reactivity to phenylephrine, possibly as a result of reduced NO bioavailability due to increased release of ROS. The increased release of ROS derived from NADPH oxidase might be due to the increased local release of angiotensin II and the release of prostanoid vasoconstrictors derived from the COX-2 pathway. Acute mercury exposure, occurring time to time could induce inflammation, vascular injury and atherosclerosis due to endothelial oxidative stress and contributing to increase peripheral resistance. Results also suggest that time to time acute exposure even at low mercury concentration might be a high risk factor for public health.