PGE 2 and 6-Keto-PGF 1α Generation and Cyclic AMP in the Atria of Rats during Normoxia and Hypoxia

Circulating antibodies can be detected in chronic periodontitis patients given the presence of serum auto antibodies against β 1 -adrenoceptor (β 1 -AR) in periodontitis patients by using cardiac membranes or a synthetic β 1 -AR peptide corresponding to the second extracellular loop of human β 1 -AR as antigens. Periodontitis patients also exhibit enhanced atria contractility in normoxia and tonic contraction in hypoxia. Atenolol, synthetic β 1 peptide and nifedipine abrogate the action of both Isoproterenol and β 1 IgG on atria contractility in both experimental conditions. In turn, β 1 IgG display a partial agonist-like activity and modifies the contractility of isolated atria, accompanied by an accumulation of cyclic AMP nucleotide in normoxia and hypoxia. The autoantibody is able to provoke an increment in the concentrations of PGE 2 and 6-keto-PGF 1α , which is abrogated by preincubating atria with adrenergic antagonist, synthetic β 1 peptide and prostanoid receptor antagonists. On this basis this study seeks to correlate the periodontitis infection to an increased risk of cardiac disease high lightening the role of β 1 IgG in the alteration of myocardial contractility and the subsequent increment of prostaglandins (PGE 2 and 6-keto-PGF 1α ,) accompanied by an increment in the production of cyclic AMP, resulting in an effective adaptation of myocardium function in acute ischemia.


Introduction
Intermittent hypoxia has been demonstrated to have powerful cardiovascular protective capabilities [1]. These effects can be grouped into three major categories: 1: adaptation of organs and tissues responsible for oxygen uptake and transport [2], 2: proliferation and increased density of vascular networks [3], and 3: increased mitochondrial density in the brain, the liver and the heart [4]. Consequently, adaptation to intermittent hypoxia has been used to treat patients with ischemic heart disease [5] and with post myocardial infarction heart failure [6]. Periodontitis is on its part characterized by a gingival inflammation in which periodontopathic bacteria generate immunological inflammatory responses and is seen as a key risk factor for the onset of cardiovascular disease [7][8][9]. In this connection it is important to mention, that autoantibodies against atria cardiac β 1 -adrenoreceptors (AR) are able to mimic the effect of an authentic β 1 -AR agonist acting on atria β 1 -AR [10] in the sera of patients with periodontitis. The joined release of host-derived inflammatory mediators (such as cytokines) and serum anti-β 1 -AR IgG into the circulation of chronically-inflamed periodontal tissues may thus provide a link between periodontal disease and cardiovascular disease [8,9]. Other studies state that the exposure of the heart to a sub lethal ischemic stress induces a phenotypic change that renders the myocardium relatively resistant to a subsequent ischemic insult occurring 24 to 72 hours later, corresponding to late phase of ischemic preconditioning [11,12]. Late phase of ischemic preconditioning is itself associated with the up regulation of cyclooxygenase-2 (COX-2) expression; the COX-2-dependent synthesis of prostanoids (ie, prostaglandin E 2 and prostaglandin I 2 ) is viewed as essential for the observed cardio protective effects [13]. Prostanoids have been shown to alleviate myocardial ischemia/reperfusion injury [14]. What remains unknown on the one hand is which are the signaling pathways whereby a sub lethal ischemic stress leads to an increased expression of prostanoids in the heart; on the other, how the modulation of COX-2 in adult myocardium takes place.

Advances in Dentistry and Oral Health
And yet, COX-2 has been shown to be unregulated by herbal ester and oxidative stress in isolated neonatal myocyte [15]. The expression of COX-2 has been reported to be controlled, among other factor(s), by reactive oxygen species (ROS) [16], protein kinase C (PKC) [17], protein tyrosine kinases (PTKs) [18] in macrophage and platelet cells and to be accompanied by an increment of PGE 2 and PGI 2 . In turn, these signaling elements are acknowledged to be implicated in late phase of ischemic preconditioning and in ischemic heart by demonstrating that the development of delayed cardio protection is triggered by the formation of nitric oxide (NO) and ROS during the initial phase of ischemic preconditioning stimulus and by the subsequent early activation of a araquidonic acid cascade that involves an increase of PGE 2 and PGI 2 with the sequential recruitment of PKC [19,20]. Consequently, in the present study, we test the hypothesis that β 1 IgG participate in the induction of COX-2 in response to sub lethal myocardial ischemia, whereas the cardiac synthesis of prostanoids such as PGE 2 and 6-keto-PGF 1α are able to protect the myocardium in the course of ischemia under normoxia and hypoxia conditions. We also study the contractile pattern of the atrium in these experimental conditions without addition of (control), and in the presence of isoproterenol and β 1 IgG. Our results suggest that the enhancement of PGE 2 and 6-keto-PGF 1α with the subsequent increment on cyclic AMP may be involved in the pathological process during hypoxia condition in isolated rat atria myocardium.

Patients
The study group consisted of 16 male adult patients with periodontitis who were attending the Periodontology Clinic from the metropolitan area of Buenos Aires. These men were aged 32 to 50 years old. Healthy subjects aged 30 to 46 years old were used as control (14 male subjects). The assessment of clinical parameters was carried out by a calibrated periodontist following on the basis of clinical parameters, attending to the severity of the disease and the periodontal tissue destruction [21]. The characteristic clinical signs of periodontitis were the following: loss of clinical attachment; horizontal or/and angular alveolar bone loss; periodontal pocket formation; and gingival inflammation. At least six sites with ongoing periodontal disease were required to be considered in the study. Clinical measurements on patients with periodontitis included sites with alveolar bone loss >2 mm and a pocket depth > 5mm with bleeding and attachment loss >3 mm. In the healthy subjects (control group), the probing depth was < 3 mm and the attachment loss was < 2 mm. The probing pocket depth and the clinical attachment level were also assessed at six sites per tooth and the bleeding on probing was assessed at four sites per tooth. None of the subjects (patients in group I and controls in group II) had either systemic illnesses or were smokers. The patients with periodontitis had not received periodontal treatment or antibiotics within the preceding 5 months or any anti-inflammatory drugs 3 weeks prior to the study. The pocket probing depth (PPD: ≥ 5 mm), the clinical attachment loss (CAL: ≥ 4 mm) and bleeding on probing: positive, are also determined. All of the patients consented to participate in the study and the investigation was conducted according to the tenets of the Declaration of Helsinki of 1975 as revised in 2000.

Human sera and IgG purification
Sera and the corresponding IgG were obtained from patients with periodontitis and healthy individuals. Six ml of blood was obtained by venipuncture and allowed to clot at room temperature, and the serum was separated by centrifugation at 2000 g and stored at -20ºC until used in assays. The IgG was obtained by precipitation with 50% ammonium sulphate and followed by 3 washes and re-precipitation with 33% ammonium sulphate. The resulting precipitate was submitted to chromatography on DEAE-cellulose, equilibrated with 10 mM phosphate buffer (pH 8). The eluted peaks were concentrated by ultra filtration to 10 mg protein/ml. Control immune electrophoresis with goat anti-human total serum and goat nonspecific anti-human IgG showed only one precipitin line.

Animals
Adult male Wistar strain rats (250-300 g) were used. The animals were housed in standard environmental conditions and fed with a commercial pellet diet and water ad libitum. The experimental protocol followed the Guide to The Care and Use of Experimental Animals (DHEW Publication, NIH 80-23). Rats were anesthetized with a mixture of ketamine and xylazine (50 and 5 mg kg -1 respectively) and killed by decapitation.

Contractile study
Rats were decapitated and atria were removed quickly and placed in a glass chamber containing a Krebs Ringer bicarbonate (KRB) solution (pH 7.4) gassed with 5% CO 2 in oxygen at 37ºC. After a stabilization period of 30 min, spontaneous tension and frequency were recorded using a force transducer coupled to an ink-writing oscillograph, as previously described [22]. Then, the preparations were paced by means of a bipolar electrode using a SK4 Grass stimulator, with stimuli duration of 2 msec and a voltage that was 10% above threshold. The constant resting tension applied to the atria (preload tension) was 750 mg. The contractility (dF/dt) was assessed by recording the maximum rate of isometric force development above the externally applied resting tension. To obtain the maximum IgG effect, different concentrations of IgG were added to normal rat atria every 10 min. The average control (basal) force of tonic contraction in equilibrium before the addition of drugs or β 1 IgG was 205 ± 18 mg, n= 10. Control values (equal to 100%) referred to the dF/dt before the addition of different IgG concentrations.

Assay for cyclic adenosine monophosphate (cAMP)
Rat atria (10 mg) were incubated in 1 ml KRB for 30 min and the β 1 -IgG added in the last 15 min in normoxia and hypoxia. When blockers were used, they were added 25 min before the Advances in Dentistry and Oral Health addition of the antibody. After incubation, atria was homogenized in 2 ml of absolute ethanol and centrifuged at 6000 × g for 15 min at 4°C. Pellets were then re-homogenized in ethanol-water (2:1). Supernatants were collected and evaporated to dryness. Cyclic AMP in the residue was dissolved in a 400 µl of 0.05 M sodium acetate buffer (pH 6.2). The determination of the nucleotide was done using ELISA according to the protocol of production of cAMP from Amersham Biosciences (Piscataway, NJ, USA). Results were expressed in Pico moles per milliliters (pmol/ml).

PGE2 and 6-keto-PGF1α assays
Atria were incubated in 0.50 ml KRB gassed with 5% CO 2 in oxygen at 37°C in normoxia and hypoxia. Concentration-response curve of β 1 IgG and isoproterenol were added to the isolated atria before the end of the incubation period. Blockers were added 10 min before the addition of different concentrations of β 1 IgG (maximal effect 1×10 -8 M) and isoproterenol (maximal effect 1×10 -7 M). Atria were then homogenized and transferred into 1.5 ml polypropylene micro-centrifuge tubes. Subsequently, all samples were processed according to the protocol supplied with the PGE 2 and 6-keto-PGF 1α Biotrak Enzyme Immunoassay (ELISA) System (Cayman Chemical, Ann Arbor, MI). The results are expressed as nanograms per milliliters (ng/ml).

Experimental Protocol
The isolated atria underwent a 50 min stabilization under basal conditions (normoxia), during which they were equilibrated in a buffer gassed with 95% O 2 , 5% CO 2 (37ºC, pH 7.4; pO 2 > 600 mmHg). In hypoxia condition atria were equilibrated in a buffer gassed with 95% N 2 , 5% CO 2 (37ºC, pH 7.4; pO 2 > 100 mmHg) for 50 min. The adrenergic agonist (isoproterenol, 1x10 -7 M) and the antibody (β 1 IgG 1x10 -8 M) were added in the last 10 min before the beginning the hypoxic period. The total duration of this experimental design was 120 min. Contractility was then registered and processed for the biochemical analysis of the production of PGE 2 , 6-keo-PGF 1α or cAMP. All this protocol was done in atria subjected to normoxia and hypoxia. In the blocking experiments, atenolol (1×10 -7 M, a β 1 -specific adrenergic antagonist) and synthetic β 1 adrenergic peptide (5×10 -5 M) were added 20 min before agonists. Normal β 1 IgG was used as control (basal values). The contractility of atria (dF/dt g/s) and the determination of tonic contraction (mg) were done in both experimental conditions (normoxia and hypoxia).

Drugs
Stock solutions of isoproterenol (ISO), atenolol, PGE 2 , 6-keto-PGF 1α , nifedipine, (Sigma Chemical Company, St. Louis, MO, USA) were freshly prepared before each experiment in their respective specific buffers. The β 1 synthetic peptide sequence corresponding to the second extracellular loop of the human β 1 -AR was HWWRA ESDEA RRCYN DPKCC DFVTN RC.

Statistical Analysis
A Student's t-test for unpaired values was used to determine the levels of significance. Analysis of variance (ANOVA) and a post hoc test (Dunnett´s method and Student-Newman-Keuls test) were employed when pair-wise multiple comparison procedures were necessary. The differences between means were considered significant at a P < 0.05.  Figure 1 shows that iso proterenol (ISO) 1x10 -7 M (maximal stimulation) enhances atria contractility through the activation of atria β 1 adrenoreceptor in normoxia, given that in the presence of the β 1 specific antagonist atenolol 1x10 -7 M, this stimulatory action of ISO was blocked. Furthermore, the β 1 IgG (1x10 -8 M) from the serum of patients with chronic periodontitis also causes a stimulation of atria contractility, which is abolished when the β 1 IgG is previously incubated with the β 1 synthetic peptide (1x10 -5 M). On the other hand when the preparation was in hypoxia, the presence of ISO at 1x10 -7 M resulted in an increase in tone. This effect was mimicked by the autoantibody β 1 IgG at 1x10 -8 M. The increment in tone, produced by both ISO

Advances in Dentistry and Oral Health
and β 1 IgG, were blunted by nifedipine (calcium blocker agent) at 5x10 -5 M. The heart is a major site of β 1 adrenoreceptors' expression and cardiac β 1 adrenoreceptors mediate the organ's response to changes in circulating nor epinephrine levels and increased myocyte contractility, resulting in positive inotropic and chronotropic effects on the heart. Figure 2A shows that in a concentration-response curve ISO enhance cardiac contractility (dF/dt) through the activation of β 1 adrenoreceptors during normoxia whereas atenolol (1x10 -7 M) blocks this action. Atenolol alone at this concentration is without effect in the system. Figure 1B shows that increasing concentrations of β 1 IgG were applied to beating isolated rat atria during normoxia, increased atria contractility, which itself reaches the maximal effect at 1x10 -8 M. This effect is almost blocked by pretreating atria with atenolol and with the β 1 synthetic peptide. Normal IgG from healthy individuals used as control did not have any effect on our preparations. During hypoxia, both ISO (1x10 -7 ) and β 1 IgG (1x10 -8 M) resulted in a time-dependent increase in tone ( Figure 1C). Both actions were abolished by preincubating atria with nifedipine (5x10 -5 M). Normal IgG was ineffective in the system.

Biochemical determinations
It can be seen in (Figure 3) that different concentrations of β 1 IgG are able to provoke an increased production of PGE 2 (upper panel, normoxia) and 6-keto-PGF 1α (lower panel, normoxia) in rat isolated atria. This action is inhibited by specific β 1 receptor antagonist atenolol 1x10 -7 M and β 1 synthetic peptide 1x10 -5 M respectively. The same (Figure 3) shows that in the course of experimental hypoxia (see upper and lower panel), the β 1 IgG significantly enhances both prostanoids and this action is blunted by atenolol 1x10 -7 M and β 1 synthetic peptide 1x10 -5 M as described above. It is interesting to note, that the increment of PGE 2 is significantly higher in hypoxia as compared to normoxia, whereas 6-keto-PGF 1α values are not significantly different. On the contrary, ISO 1x10 -7 upon rat isolated atria is able to increase the generation of PGE 2 and 6-keto-PGF 1α in both experimental conditions (normoxia versus hypoxia), and atenolol blunted this action. Normal IgG is unable to modify the generation of both prostanoids in our system ( Table 1). As shown in (Table 2), β 1 IgG and ISO trigger an increase in the levels of endogenous cAMP in comparison with basal values of this nucleotide in rat isolated atria during normoxia and hypoxia. It is important to note, that the increment of cAMP observed in hypoxia condition, are significant lower than those observed in normoxia conditions.

Discussion
An increased prevalence of cardiovascular and autoimmune diseases in periodontitis has been reported [23,24]. Herein, we demonstrate an association between periodontal infections and an increased risk of cardiovascular disease, pointing to the role of anti β 1 -adrenoreceptor (AR) antibodies (β 1 IgG) in the serum of patients with periodontitis. The results provide evidence that components of the serum IgG fraction from patients with chronic periodontitis recognize cardiac membranes. Most of the sera that react positively against the surface of rat cardiac membranes show a positive immune reactivity to the human β 1 -AR peptide. In this sense, we establish that the β 1 -AR is the target for the anti-rat cardiac auto-antibodies described in patients with periodontitis [25] using a synthetic peptide with an identical amino acid sequence of the second extracellular loop of cardiac human β 1 -AR. Knowing that the amino acid sequence of rodent and human β 1 -AR peptides have strong homology [26], we study the β 1 IgG mediated effect of auto-antibodies from patients with periodontitis on rat atria tissue contractility. It is generally accepted that the heart is a major site of β 1 -AR expression when it mediates the positive chronotropic effect on the heart. Given that the serum of patients with periodontitis contains autoantibodies against the cardiac and human gingival fibroblast membrane β 1 -AR, we tested the atria contractility in the presence of an authentic β 1 adrenergic agonist isoproterenol and β 1 IgG from patients with periodontitis. The results in normoxia condition provide evidence that components of the serum β 1 IgG fraction from patients with periodontal disease enhance atria dF/dt in that way ISO does, i.e., by recognizing and activating the atria β 1 -AR. This is the case since atenolol and β 1 synthetic peptide inhibited the stimulatory action of the antibody upon dF/dt on rat atria contractility, we study the β 1 -AR-mediated Advances in Dentistry and Oral Health effect of auto-antibodies (β 1 IgG) from patients with periodontitis on rat cardiac tissue. Now when rat atria is exposed to hypoxia, ISO and β 1 IgG are only able to provoke a tonic contraction without rhythmic activity, which is in turn depressed by nifedipine, an calcium channel inhibitor. It may be that there is an alteration in the kinetic of myocardium intracellular calcium concentration handling in hypoxia condition and this is what provokes an altered calcium flux as nifedipine abrogates the tonic contraction. A similar phenomenon was described in infarcted heart [27]. Nifedipine was also considered more potent blocking calcium channel agents as an inhibitor of depolarizationinduced contractions of arteries [28]. We also demonstrated in previous works that those patients from whom sera contain anti-cardiac and anti β 1 IgG have a decrease in heart rate variability (HRV). This is an important finding with respect to the in vivo effects of the auto-antibodies, since the aberrant β 1 sympathetic receptor's expression may result in a decrease in HRV and it is a risk factor for the development of cardiovascular diseases, including heart failure, myocardial infarction, and hypertension [29]. Experimental hypoxia is associated with the changes of the PGE 2 /6-keto-PGF 1α systems. Considering that PGE 2 and PGI 2 are mainly located in the cytosolic supernatant fraction of the heart atria homogenates and arteries, and their concentrations (ng/mg protein) are significantly higher in atria tissue than in ventricles, this association supports the idea of a regional variability in the araquidonic acid cascade and its products derived from the cyclooxygenase(s) system during hypoxia. Alterations in the arachidonic acid's metabolism are often involved in the myocardial disturbances associated with hypoxia and ischemia [30]. Additionally [31] demonstrated that the prostanoid production pattern changes in hypoxic rat atria, generating a cardioprotective action involving the prevention of the increases of lactate and cAMP during ischemia [32]. If so, then PGE 2 and PGI 2 would be expected to have an important physiological role in the hypoxic rat heart. In this sense, a possible mechanism could be that just after the hypoxic stimulus, COX-2-derived PGE 2 and 6-keto-PGF 1α generation rises in our study preparations exerting a vasodilatory effect in order to regulate and/or modulate the maintenance of blood flow through the hypoxic tissues. Thus, the maintained levels observed in PGE 2's and 6-keto-PGF 1α's systems during hypoxia could be involved in the adaptation of the heart to such situations. Hypoxia seems to directly stimulate the expression of atria cox-2 enzyme, provoking a specific enhancement in the synthesis of PGE 2 and 6-keto PGF 1α. The result is vasodilatation and a decreased vascular resistance of the ischemic bed and the subsequent improvement of the blood flow that in turn, facilitates cardiac function [33]. Atenolol abrogates the effects of both ISO and the β 1 IgG but the synthetic β 1 peptide only abrogates the effect of the autoantibodies. This demonstrates that these autoantibodies recognize the cardiac β 1 adrenoceptor from sarcolemma of rat atria and its capacity to interact with atria β 1 -AR. Moreover, acting as inducers of cox-2 mRNA levels, in turn, provoked an increase in the cardiac proinflammatory substances such as PGE 2 and PGI 2 . The major new finding of this work is the demonstration that an anti-β 1 adrenoceptor IgG behaving as a partial adrenergic agonist, has the capacity to alter the rate of transcription of specific proinflammatory target genes, triggering the cardiac production of PGE 2 and 6-keto PGF 1α in response to receptor-mediated signaling events at the receptor cell cardiac atrium. The coupling of the β 1 IgG and ISO on β 1 -AR in atria permits the transduction of one signalling pathway as cAMP elevation. This effect was abrogated by an antagonistic β 1 -AR agent (atenolol) and by the synthetic β 1 peptide. It is interesting that the partial agonist activity of the auto-antibodies observed in vitro mimicked the action on HRV of a partial β 1 adrenergic agonist (celiprolol) described in vivo [34]. In our work, ISO and β 1 IgG have the capacity to increase the levels of cAMP at similar values, whereas in hypoxia condition, both the levels of cAMP and the contractile force (tonic contraction) were lower than in normoxia. These results indicate changes in the prostanoids' generation associated with changes in the adenylate cyclase activity and the physiological contractile function of the rat atria. On the basis of our results we suggest the possibility that the alterations in the contractile pattern of atria in the presence of ISO and β 1 IgG be caused by antibody fixation and activation of the cardiac β 1 -AR. If so, then, β 1 IgG could exacerbate the course of ischemic heart diseases by altering the focus of immune function with up-regulation of its own production and suppression of cell mediated immunity. It is important to note, that patients with periodontitis might develop the anti β 1 -AR autoantibodies (β 1 IgG) as a result of molecular mimicry by bacterial pathogens acting on cardiac β 1 -AR as a result of receptor alteration and/or degradation during the inflammatory response. The microbial challenge and inflammatory response in the periodontium via a similar mechanism could be causing increased risk for myocardium diseases. This being the case, neo-antigens from cardiac β 1 -AR tissue may be the key to generation of these autoantibodies (β 1 IgG) coupled to an increased generation of prostanoids and cyclic AMP accumulation, i.e. a clinical manifestation of autoimmunity. The pathogenic properties of anti β 1 IgG antibodies have been ascribed to their potency to continuously stimulate the sympathetic system. Thus, β 1 IgG triggers the following two important mechanisms: the generation of autoantibodies by targeting cardiac β 1 -AR altering the physiologic behaviour of the myocardium; the exacerbating or maintaining of a chronic inflammatory process in periodontitis disease through the induction of a suppressed immune response [35,36] and an enhancement of PGE2's and 6-keto PGF 1α's production. Cardiac eicosanoid's release in chronic hypoxic conditions is primarily beneficial and results in an effective adaptation of the heart at its function. In acute ischemia it is however primarily deleterious, since it enhances inflammatory reactions with an increased generation of prostanoids, different cytokines and ROS molecules.

Advances in Dentistry and Oral Health Conclusion
On the other hand, the autoantibody provokes an increment in the concentrations of PGE 2 and 6-keto-PGF 1α that is abrogated by the pre incubation of atria with the adrenergic antagonist, and the synthetic β 1 peptide respectively. We demonstrate in this study an association between periodontitis infection and an increased risk of cardiac disease, highlighting the role of β 1 IgG in the alteration of myocardial contractility with an increment of prostaglandins (E 2 /I 2 ) turnover together with an increment in the production of cyclic AMP, resulting in an effective adaptation of the myocardium's function in acute ischemia.