The role of potassium channels in the endothelial dysfunction induced by periodontitis

Abstract Objective: Periodontitis is associated with endothelial dysfunction, which is clinically characterized by a reduction in endothelium-dependent relaxation. However, we have previously shown that impairment in endothelium-dependent relaxation is transient. Therefore, we evaluated which mediators are involved in endothelium-dependent relaxation recovery. Material and methods: Rats were subjected to ligature-induced experimental periodontitis. Twenty-one days after the procedure, the animals were prepared for blood pressure recording, and the responses to acetylcholine or sodium nitroprusside were obtained before and 30 minutes after injection of a nitric oxide synthase inhibitor (L-NAME), cyclooxygenase inhibitor (Indomethacin, SC-550 and NS- 398), or calcium-dependent potassium channel blockers (apamin plus TRAM- 34). The maxilla and mandible were removed for bone loss analysis. Blood and gingivae were obtained for C-reactive protein (CRP) and myeloperoxidase (MPO) measurement, respectively. Results: Experimental periodontitis induces bone loss and an increase in the gingival MPO and plasmatic CRP. Periodontitis also reduced endothelium-dependent vasodilation, a hallmark of endothelial dysfunction, 14 days after the procedure. However, the response was restored at day 21. We found that endothelium-dependent vasodilation at day 21 in ligature animals was mediated, at least in part, by the activation of endothelial calcium-activated potassium channels. Conclusions: Periodontitis induces impairment in endothelial-dependent relaxation; this impairment recovers, even in the presence of periodontitis. The recovery is mediated by the activation of endothelial calcium-activated potassium channels in ligature animals. Although important for maintenance of vascular homeostasis, this effect could mask the lack of NO, which has other beneficial properties.


Introduction
Periodontitis is a chronic inflammatory disease that compromises the integrity of tooth-supporting tissues. In addition to the local effects of this disease, periodontitis is also associated with systemic inflammation. 2,17 Furthermore, periodontitis has been associated with endothelial dysfunction, 15,23,33 which is an early event in the development of cardiovascular diseases, especially atherosclerosis. 34 One of the hallmarks of endothelial dysfunction is a reduced response to endothelial-dependent stimuli, such as acetylcholine. Acetylcholine stimulates endothelial nitric oxide synthase (NOS-3) to generate nitric oxide (NO), which diffuses to the underlying smooth muscle cells, inducing relaxation by increasing the production of cGMP, leading to a transient reduction in blood pressure in vivo. 2 Thus, endothelial dysfunction is broadly defined as an impairment in vascular relaxation, due to decreased NO production by the endothelium and/or increased inactivation of NO. 14 Besides being a potent vasodilator, NO has antithrombotic, anti-inflammatory, and antimitogenic properties, which explains why the reduction in NO levels is associated with increased cardiovascular disease risk. 21 However, the perception of endothelial dysfunction as just a reduction in NO production/bioavailability is oversimplified. In vitro experiments have shown that, in large conduit vessels such as the aorta, acetylcholine-induced vasodilatation is predominantly mediated by NO. 29 However, in addition to NO, other mediators such as prostacyclin (PGI 2 ) and endothelium-dependent hyperpolarization (EDH) contribute to endothelium-dependent vasodilator response to agonists in small resistance vessels. 1 Prostacyclin is a product of the metabolism of arachidonic acid by cyclooxygenase (COX), and acts through the prostacyclin receptor (IP). 18 EDH has been proposed to mediate vasodilation through the initial activation of small conductance (K Ca 2.3) and intermediate conductance (K Ca 3.1) calcium-activated potassium channels, which are present on the endothelium. 10 Following the opening of the K Ca 2.3 and K Ca 3.1 channels in the endothelial cell, vascular smooth muscle cell hyperpolarization is evoked by electrical coupling through the myoendothelial gap junction ( Figure 1). 7 Interestingly, a compensatory increase in EDH and/or prostacyclin-mediated vasodilation in response to acetylcholine has been demonstrated in blood vessels of NOS-3 knockout mice. 12,19,32 Therefore, there is a redundancy in the system, and more than one endothelial mediator is capable of acting as the signal between endothelium and smooth muscle. This could explain why some studies have reported normal endothelium-dependent relaxation in models of atherosclerosis. 25 We have previously demonstrated a reduction in acetylcholine response 14 days after ligature-induced periodontitis. 2 However, despite lasting systemic and vascular inflammation, impairment in the acetylcholine response is restored 21 days onwards after the procedure. 2,24 Since endothelium-dependent relaxation is the main mechanism for assessing endothelial dysfunction, 5 the compensatory effect could mask a lack of NO and an increased risk of cardiovascular disease.
Therefore, the main aim of this study was to assess the involvement of nitric oxide, potassium channels, and COX products in the restoration of the endothelium-dependent response during periodontitis.  Eighteen animals were subjected to a protocol similar to the one previously described; however, all of them were subjected to the ligature procedure.
Twenty-one days after the procedure, the animals were prepared for blood pressure measurement. After  Figure 3A). However, the response to acetylcholine was similar to day 0 in rats submitted to ligature 21 days earlier ( Figure 3A). In contrast, the response to sodium nitroprusside was similar among the groups at all evaluated times ( Figure 3B). There is no change in basal blood pressure in the times evaluated ( Figure   3C).

Involvement of NOS, COX, or EDH restoration of the acetylcholine-induced reduction in blood pressure
Because the main objective of this study was to evaluate the mechanism involved in the restoration OLCHANHESKI JR LR, SORDI R, OLIVEIRA JG, ALVES GF, MENDES RT, FERNANDES D Bone loss of the maxilla and mandible was measured on day 0, 14 days, or 21 days after the ligature procedure (A). Bone loss was quantified as the distance between the cement enamel junction and the alveolar crest (in mm). Myeloperoxidase (MPO) activity was also measured in the gingival tissue around the ligated molars (B). Blood was collected and assayed for C-reactive protein (CRP) (C). Statistical analysis was performed using one-way analysis of variance followed by Bonferroni post hoc test. *p<0.05, **p<0.01, and ***p<0.001 2018;26:e20180048 6/10 of acetylcholine-induced reduction in blood pressure, the next experiments were performed only 21 days after the ligature.
The administration of the NOS inhibitor L-NAME in sham or periodontitis rats did not change the peak reduction in mean arterial pressure to sodium nitroprusside or acetylcholine ( Figure 4A-B). Similarly, the non-selective COX inhibitor, indomethacin, did not change the sodium nitroprusside or acetylcholine response of sham or periodontitis rats ( Figure 4C-D).

Discussion
In this study, we observed the development of a compensatory endothelium-dependent hypotension in animals with experimental periodontitis, which was, at least in part, mediated by the opening of calciumactivated potassium channels. These findings suggest that endothelial dysfunction is a temporally complex event.
Disturbances in endothelial function leading to reduced endothelium-dependent vasodilation are present in patients with cardiovascular risk factors, including hypertension, hypercholesterolemia, diabetes, smoking, ageing, atherosclerotic disease, and inflammation. 6 Moreover, periodontitis has also been associated with endothelial dysfunction. 15,33 Ligature-induced periodontitis in rats is among the most widely used experimental models of periodontitis. Alveolar bone loss, which is the main characteristic of periodontitis, is well-established seven days after ligature placement; 2 however, previous work from our group has demonstrated that systemic and vascular inflammation is more consistent from 14 days on. 2,24 Therefore, we performed our analysis two and three weeks after ligature placement. Two weeks after ligature placement, progressive bone The role of potassium channels in the endothelial dysfunction induced by periodontitis Figure 3-Transient endothelial dysfunction induced by periodontitis. Ligature rats were prepared for mean arterial pressure recording 1 h (day 0), 14 days, or 21 days after the procedure. Increasing doses of acetylcholine (A) or sodium nitroprusside (B) were injected, and the change in mean arterial pressure (% of decrease from basal blood pressure) was determined. The basal mean arterial pressure is also shown (C). Statistical analysis was performed using one-way analysis of variance followed by Bonferroni post hoc test. **p<0.01 and *** p<0.001 2018;26:e20180048 7/10 loss was observed, suggesting that local inflammation was not resolved even after three weeks. This is corroborated by the increased levels of MPO two and three weeks after ligature placement. Two weeks after ligature placement, there was also endothelial dysfunction, agreeing with our previous results 2 and with human studies. 15,33 However, the reduced endothelium-dependent relaxation, which has been used as an indicator of endothelium function, was transient, and normal relaxation was present three weeks after ligature placement. 24 Therefore we have confirmed previous results showing that impairment OLCHANHESKI JR LR, SORDI R, OLIVEIRA JG, ALVES GF, MENDES RT, FERNANDES D Figure 4-Involvement of NOS, COX, or EDH restoration of the acetylcholine-induced reduction in blood pressure. Twenty-one days after ligature or sham procedure, the animals were prepared for measurement of mean arterial pressure, and dose response curves to acetylcholine and sodium nitroprusside were performed before and 30 min after L-NAME (20 mg/kg, i.v; panels A-B), indomethacin (10 mg/kg, i.v; panels C-D) or apamin plus TRAM-34 (75 µg/kg and 1 mg/kg, respectively, i.v; panels E-F). The change in mean arterial pressure (% of decrease from basal blood pressure) was determined. Statistical analyses were performed using two-way analysis of variance followed by Bonferroni post hoc test. *p<0. However, in some cases, particularly in small resistance type vessels, other mediators besides NO, such as EDH and/or PGI 2 , contribute to endothelium-dependent vasodilator response to agonists. 7 In endothelial cells, PGI 2 is the main COX product. Prostacyclin is a potent vasodilator and its effect is mainly mediated through specific cell surface receptors, known as IP. 18 EDH is activated by an increase in intracellular calcium concentration in endothelial cells, followed by the opening of K Ca 2.3 and K Ca 3.1 channels, and the subsequent hyperpolarization of these cells (Figure 1).
The resulting endothelial hyperpolarization spreads via myoendothelial gap junctions, which then results in the hyperpolarization and relaxation of the smooth muscle. 7,10 With this in mind, we explored the role of these endothelium relaxant pathways in restoring acetylcholine response. It has been demonstrated that, in diabetic rats, despite increased oxidant stress, endothelium-dependent relaxation is maintained due to increased NOS-3 expression, resulting in increased NO synthesis. 16 Additionally, an increase in NOS-2 (iNOS) expression in the aorta of rats with periodontitis has recently been demonstrated. 3 However, given that the non-selective NOS inhibitor failed to change