j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Loss of vagal tone aggravates systemic inflammation and cardiac impairment in endotoxemic rats Astrid Schulte, MSc, Christoph Lichtenstern, MD,* Michael Henrich, DPhil, Markus A. Weigand, MD, and Florian Uhle, MSc Department of Anesthesiology and Intensive Care Medicine, Justus-Liebig-University Giessen, Giessen, Germany
article info
abstract
Article history:
Background: During the course of sepsis, often myocardial depression with hemodynamic
Received 17 September 2013
impairment occurs. Acetylcholine, the main transmitter of the parasympathetic Nervus
Received in revised form
vagus, has been shown to be of importance for the transmission of signals within the
13 January 2014
immune system and also for a variety of other functions throughout the organism. Hy-
Accepted 16 January 2014
pothesizing a potential correlation between this dysfunction and hemodynamic impair-
Available online 21 January 2014
ment, we wanted to assess the impact of vagal stimulation on myocardial inflammation and function in a rat model of lipopolysaccharide (LPS)-induced septic shock. As the
Keywords:
myocardial tissue is (sparsely) innervated by the N. vagus, there might be an important
Septic cardiomyopathy
anti-inflammatory effect in the heart, inhibiting proinflammatory gene expression in
Endotoxemia
cardiomyocytes and improving cardiac function.
Vagus nerve stimulation
Materials and methods: We performed stimulation of the right cervical branch of the N. vagus
Acetylcholine
in vagotomized, endotoxemic (1 mg/kg body weight LPS, intravenously) rats. Hemody-
Cholinergic antiinflammatory reflex
namic parameters were assessed over time using a left ventricular pressureevolume catheter. After the experiments, hearts and blood plasma were collected, and the expression of proinflammatory cytokines was measured using quantitative reverse transcription polymerase chain reaction and enzyme-linked immunosorbent assay. Results: After vagotomy, the inflammatory response was aggravated, measurable by elevated cytokine levels in plasma and ventricular tissue. In concordance, cardiac impairment during septic shock was pronounced in these animals. To reverse both hemodynamic and immunologic effects of diminished vagal tone, even a brief stimulation of the N. vagus was enough during initial LPS infusion. Conclusions: Overall, the N. vagus might play a major role in maintaining hemodynamic stability and cardiac immune homeostasis during septic shock. ª 2014 Elsevier Inc. All rights reserved.
1.
Introduction
Sepsis and septic shock are, despite intensive research and new therapeutic options, the major causes of death in
intensive care units in Western civilization [1e4]. Sepsisinduced cardiac dysfunction often occurs during the course of disease and is characterized by hemodynamic and cardiovascular instability [5e7] accompanied by increased mortality
* Corresponding author. Department of Anesthesiology and Intensive Care Medicine, Justus-Liebig-University Giessen, Rudolf-Buchheim-Straße 7, 35392 Giessen, Germany. Tel.: þ49 641 985 44401; fax: þ49 641 985 44409. E-mail address: [email protected] (C. Lichtenstern). 0022-4804/$ e see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.01.022
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
[8]. Besides occurring inflammatory processes [9], one important aspect in septic cardiomyopathy is the failure of parasympathetic innervation, leading to autonomic dysfunction with, for example, reduced variability of the heart rate [10]. At the beginning of the 20th century, Otto Loewi [11] published his results of the “Vagusstoff” experiments. He showed that the “Vagusstoff”, later identified as Acetylcholine (ACh) by Henry Dale [12], is secreted from isolated, vagus nervee stimulated frog hearts. Approximately 80 y later, it was shown that ACh, the principal vagal neurotransmitter, was able to reduce proinflammatory cytokine levels in response to endotoxin in vitro in human macrophages and in vivo through direct electrical stimulation of the efferent vagal nerve in rats [13]. These experiments highlighted the link between immune and vegetative nerve system, and the idea about the “cholinergic anti-inflammatory reflex” was postulated [14]. First, it was assumed that the spleen, innervated by catecholaminergic splenic nerve fibers originating in the celiac-superior mesenteric plexus ganglia, is required for the cytokine regulatory effect [15]. However, a recent study provided evidence that the vagal preganglionic terminals do not synapse with the sympathetic neurons innervating the spleen [16]. In this context, the question raised about a “nonneuronal cardiac cholinergic system” [17] with an influence of cardiac innervation during inflammation, introducing the heart as the possible interface between immune and nervous system. Analysis of cardiac myocytes indicated that these cells express the specific ACh receptor subunit for mediating this reflex [18] and, furthermore, are able to synthesize the enzymes necessary for producing ACh and to react in a paracrine fashion [19]. After Otto Loewi’s approach of vagal stimulating hearts, but staying in the living animal, we performed hemodynamic measurements with stimulation of the right cervical branch of Nervus vagus from Lewis rats under LPS-induced septic shock to prove our hypothesis that the N. vagusederived ACh, released from terminal endings inside the cardiac tissue on stimulation, inhibits the proinflammatory gene expression in cardiomyocytes and improves cardiac function during septic cardiomyopathy.
2.
Material and methods
2.1.
Animal handling and care
Male Lewis rats (275e300 g) were purchased from Charles River (Sulzfeld, Germany). Animals were kept in cages with free access to water and food on a 12-h light and dark cycle. All procedures involving animals conform to the standards for animal experiments and were approved by the local committee for animal care (No. 108-2010, Regierungsprasidium Giessen, Germany).
2.2.
Animal model of endotoxic shock
Experiments were performed in anesthetized rats. Anesthesia was induced (5%) and maintained (2%e3%) with isoflurane (Baxter, Unterschleissheim, Germany). After endotracheal intubation with a 16-G catheter, animals were ventilated with a rodent respirator (Harvard Inspira, MA) in a weight-adjusted
481
manner. Electrocardiogram electrodes were placed, and the heart rate was recorded. Body temperature was rectally monitored and maintained around 37 C by a heating pad (Harvard Apparatus, MA) and a red light heating lamp. Ringer solution (Braun, Melsungen, Germany; 10 mL/kg/h) and Fentanyl (Ratiopharm, Ulm, Germany; 10 mg/kg/h) were injected intravenously over the lateral tail vein through a syringe pump (Braun). A Micro-Tip catheter (SPR-1000; Millar Instruments, Houston, TX) was inserted in the tail artery for continuous measurement of the arterial blood pressure. Animals were divided in a right cervical vagotomy group with no stimulation (vagotomized [VGX] þ LPS, n ¼ 5), short electrical stimulation (VGX þ LPS þ 2 min stim, n ¼ 6), long electrical stimulation (VGX þ LPS þ 20 min stim, n ¼ 6), and two groups (sham, n ¼ 6 and sham þ LPS, n ¼ 6) without vagotomy. The right trunk of the N. vagus was gently exposed in all animals, but only transected in the VGX rats. For electrical stimulation in the VGX animals, the vagal nerve trunk was placed between the electrodes that were connected to a constant voltage stimulator (DeMeTec, Langgoens, Germany). Stimulation (5 V, 2 ms, and 1 Hz) was performed for 20 min, starting 5 min before LPS injection (t0) or for 2 min (1 min before and 1 min after LPS administration). As a result of its amphiphile structure, LPS (Escherichia coli 0111:B4, LPS-EB Ultrapure; InvivoGen, San Diego, CA; 5 mg/mL in saline) is capable of forming micelles in aqueous solution. Therefore, it was sonicated for 30 min before injection. All animals, except the sham group, received an LPS bolus (1 mg/kg in 1 mL saline, intravenously over tail vein) more than 20 min via a syringe pump (Harvard Apparatus). Experiments were finished when the animals died during the experimental protocol or 6 h after LPS or placebo (0.9 % saline) administration. After euthanasia, blood samples were collected and centrifuged, and the plasma was stored at 80 C. Hearts were excised, dissected into left and right ventricle, and flash frozen in liquid nitrogen. Storage was carried out at 80 C.
2.3.
Hemodynamic measurements
Cardiac function studies of the left ventricle were performed inserting a pressureevolume conductance catheter (SPR-838; Millar Instruments, Houston, TX), using the closed chest approach [20]. Under deep anesthesia, the right carotid artery was exposed and separated from surrounding connective tissue and the vagus nerve. The artery was punctured with a 26-G needle, and the catheter was inserted and fixed with a suture and carefully moved into the left ventricle. Preparation and catheterization were accomplished in approximately 1 h, followed by a recovery time of 30 min before the LPS bolus. Pressureevolume signals were digitized with a PowerLab 8/30 signal convertor (ADInstruments, Spechbach, Germany) and recorded using the LabChart7 Pro Software for Windows (ADInstruments). In each individual animal, parallel conductance catheter calibration was performed; therefore, 30 mL of hypertonic saline (10%) was injected through a polyethylene catheter into the right jugular vein. Once, cuvette calibration for blood conductivity determination was performed with fresh heparinized, warm blood. The evaluation of the recorded data was performed with PVAN 1.1 (Millar Instruments), and hemodynamic parameters (cardiac output [CO], ejection fraction [EF], and heartbeat) were determined.
482
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Fig. 1 e Hemodynamic measurements in control and endotoxemic rats. Heart rate (HR), (A) and blood pressure (B) decreased during 6-h anesthesia in LPS-treated rats and in control animals. LPS also impaired CO, (C) and EF, (D). Animals at risk (E). Controls are depicted by dotted lines; LPS animals are displayed by solid lines. n [ 6 at t0. Data are mean ± standard error of mean.
2.4.
Protein isolation from heart tissue
Frozen heart tissue (50e100 mg) was homogenized with a bead mill (Schwingmuehle NM 301; Retsch, Haan, Germany) in 1 mL 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid buffer (20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 1.5 mM ethylenediaminetetraacetate, pH 7.4, Protease Inhibitor [Roche, Mannheim, Germany]). After centrifugation, supernatant was collected, and total protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific, Rockford, IL).
2.5.
Cytokine enzyme-linked immunosorbent assay
Plasma samples and proteins from left and right ventricle were centrifuged, and enzyme-linked immunosorbent assay was performed using Quantikine Rat TNF-a (RTA00) and interleukin 1b (IL-1b; RLB00) kits (R&D Systems, Abingdon, UK) according to the manufacturer’s information. In case of tissue sample, cytokine levels were normalized against total protein content.
2.6. RNA Isolation, complementary DNA synthesis, and quantitative real-time polymerase chain reaction For RNA extraction, the right and left ventricles (50e100 mg) were homogenized with a bead mill (Schwingmuehle NM 301; Retsch) in 1 mL TRIzol Reagent (Ambion, Carlsbad, CA)
and then stored at room temperature for 5 min. After centrifugation, the supernatant was removed in a new tube and 0.2 mL chloroform (Roth, Karlsruhe, Germany) was added to the homogenate, mixed and allowed to stand at room temperature for 2 min. After incubation, the sample was centrifuged again at 12.000g for 15 min at 4 C. The aqueous phase was carefully removed and transferred to a new tube. An equal volume of 70% ethanol was added. Afterward, the sample was loaded on an RNeasy Mini Kit column (Qiagen, Hilden, Germany). At this step, isolation was performed according to the manufacturer’s protocol, and RNA was stored at 80 C. For analysis, 0.5 mg of RNA was reversely transcribed using the QuantiTect Reverse Transcription Kit (Qiagen). Quantitative real-time polymerase chain reaction (qRt-PCR) was performed using TaqMan Gene Expression Master Mix (Applied Biosystems, Foster City, CA). Assay numbers were as follows: Rn01453442_m1 (choline acetyltransferase [ChAT]), Rn01758585_m1 (carnitine acetyltransferase [CrAT]), Rn01525859_g1 (tumor necrosis factor alpha [TNF-a]) and Rn00667869_m1 (b-Actin). For data analysis, the comparative DCt method was used (2^[Ct ActinCt gen-of-interest]).
2.7.
Statistical analysis
All statistical analysis was performed using Graphpad Prism version 5.0f for Mac (GraphPad Software, La Jolla, CA). Group comparisons were done using the KruskaleWallis test followed by Dunn’s post-test. KaplaneMeier graphs were drawn
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
483
(369 13.5 bpm) after LPS bolus in comparison with untreated animals (353 25.8 bpm) but reaches the same level again at time point 3.5 h (LPS: 336 17.8 bpm versus sham: 335 30.7 bpm; Fig. 1A). Depression of arterial blood pressure was more distinctive in the LPS animals as in the untreated controls (Fig. 1B), which pointed to an LPS-induced effect. CO and EF were both increased after LPS application (Fig. 1C and D) but showed a strong reduction and impairment of cardiac function after 3 h of LPS administration.
3.3.
Survival time in endotoxemic, VGX rats
Survival time of VGX animals in combination with LPS injection was reduced compared with endotoxemic animals with an intact vagus nerve (Fig. 2B in comparison with Fig. 2A). All rats (n ¼ 5) from the VGX þ LPS group died within 2 h after LPS application, two died after 1 h and accordingly after 1.5 h.
3.4. Effects of electrical stimulation of the N. vagus in VGX, endotoxemic rats
Fig. 2 e Survival analysis of endotoxemic, VGX, and stimulated rats. KaplaneMeier curve of control and LPStreated animals (A). No mortality occurred in the control group, only one animal remained alive after 6 h following LPS-injection. KaplaneMeier curve of endotoxemic, VGX, and stimulated rats (B). Survival time was reduced because of vagotomy and LPS application but was improved by VNS (n ‡ 5). for visualization of survival time, and the groups were compared using the log-rank test.
3.
To determine whether the loss of vagal tone could be reestablished by constant voltage pulses, animals underwent vagotomy with vagus nerve stimulation (VNS) for 2 min (VGX þ LPS þ stim_2 min, n ¼ 6) or 20 min (VGX þ LPS þ stim_20 min, n ¼ 6). Electrical stimulation improved survival time (Fig. 2B) in VGX, LPS-induced septic shock rats compared with unstimulated, VGX controls. Heart rate was only slightly decreased; animals with short-term stimulation showed even a small increase (Fig. 3A). Arterial blood pressure was strongly reduced in unstimulated, VGX rats after LPS administration, whereas VNS effected an attenuate reduction (Fig. 3B). Hemodynamic parameters (CO, EF) also showed less impairment (Fig. 3C and D) with stimulation after LPS application, but after about 3 h, a strong decrease was observed as well.
Results
3.1. Hemodynamic measurements in rats more than 6 h under anesthesia For our prolonged hemodynamic experiments, we chose the closed chest approach for left ventricle catheterization, because of its less invasive character [20]. During a 6-h anesthesia period with isoflurane in catheterized, untreated Lewis rats (n ¼ 6), a decreased heart rate (Fig. 1A) and arterial blood pressure (Fig. 1B) could be observed over the experimental period. CO (Fig. 1C) and EF (Fig. 1D) remained constant over the whole duration of the experiments. For optimal maintenance of the vital functions and to stabilize the animals for a longer experimental period, body temperature was kept at about 37 C, and an adequate amount of fluid was substituted.
3.2. Hemodynamic measurements in rats after endotoxemia Among the LPS group (n ¼ 6), only one animal survived the whole duration of the experiment, one died after 5 h, three after 4.5 h and another one 4 h after LPS injection (Fig. 2A). Heart rate in LPS-treated rats was slightly increased at 1.5 h
3.5. Cytokine TNF-a and IL-1b levels in plasma and cardiac tissue For evaluation of the impact of vagal stimulation, cytokine levels of TNF-a and IL-1b in plasma and left and right ventricular tissues were determined using enzyme-linked immunosorbent assay. In plasma, LPS application led to a slightly increase of TNF-a and IL-1b compared with the untreated control group (Fig. 4A and B). In VGX rats, TNF-a (P < 0.001) and IL-1b levels were markedly elevated, whereas VNS moderated this increase (Fig. 4A and B). Expression patterns of TNF-a in the left and right ventricles corresponded to the patterns that were found in the plasma (Fig. 5A and B). The inflammatory response was increased by vagotomy, whereas VNS attenuated this effect. Short-term stimulation showed a trend to be more effective, but this was not statistically significant. Noteworthy, IL-1b levels in the left and right ventricles were elevated in animals receiving LPS and VGX þ LPS but were further increased in stimulated animals (Fig. 5C and D). Contrary, the highest levels of IL-1b were measured by shortterm stimulation (P < 0.01, compared with untreated controls). Remarkably, cytokine patterns of left and right ventricles differed in expression (Fig. 5).
484
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Fig. 3 e Effects of electrical VNS in VGX, endotoxemic rats. (A) Heart rate (HR), (B) blood pressure (C) CO, and (D) EF are strongly reduced in VGX, endotoxemic animals. Hemodynamic parameters were ameliorated in the vagal stimulation groups in comparison with nonstimulated rats. Animals at risk (E). Controls (VGX) are the dotted line, animals with short stimulation (VGX D 2 min) are displayed by triangle line, and animals with long stimulation (VGX D 20 min) are displayed by square line. n ‡ 5 at t0. Data are mean ± standard error of mean.
3.6. tissue
TNF-a messenger RNA expression in ventricular
To further confirm the results of inflammatory response on gene expression level in the right and left ventricles, qRT-PCR was performed. TNF-a messenger RNA (mRNA) expression was upregulated by the induction of LPS compared with untreated controls. Levels in VGX animals were significantly higher (P < 0.001 in the right ventricle). Consistent with the findings on protein level, TNF-a mRNA expression in the
stimulation groups was on a lower level. The lowest expression was found in animals with short-term stimulation, but it remained above the level measured in control animals (Fig. 6A and B).
3.7. tissue
ChAT and CrAT mRNA expression in ventricular
To determine if myocardial cells are able to produce ACh, qRTPCR was performed against ChAT, the ACh producing enzyme.
Fig. 4 e Plasma cytokine protein levels. Elevated proinflammatory TNF-a (A) and IL-1b (B) levels were measured in VGX, endotoxemic animals. Vagal stimulation decreased the plasma levels of both mediators, however, they remained above the levels of controls. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5).
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
485
Fig. 5 e Cardiac cytokine protein levels. TNF-a was increased in endotoxemic, VGX rats in the (A) left ventricle (LV) and (B) right ventricle (RV). On VNS, it was less enhanced. Levels of IL-1b protein were slightly elevated in the LPS and LPS D VGX group but to a greater extent in the stimulation groups both in left (C) and right (D) ventricular tissue. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5, except left ventricle IL-1b: n ‡ 3).
Mentionable, mRNA expression of ChAT was not detectable in any ventricular tissue of the animals. ACh can also be synthesized by CrAT. Slavikova et al. [21] have shown that ChAT activity is very low in the rat ventricles. They suspected that synthesized ACh in the ventricles is a product of CrAT. This former observation could be confirmed by qRT-PCR analysis in this study. CrAT mRNA expression was detectable in all groups with no significant regulation in the left ventricle (Fig. 7). In the right ventricle, downregulation of CrAT was observed in LPS-treated animals with the strongest reduction in VGX rats (P < 0.01, Fig. 7B). Eventually, ACh in the ventricles is synthesized by CrAT, and this enzyme is downregulated
significantly in VGX rats on LPS stimulation in the right ventricle.
4.
Discussion
The present study provides evidence that LPS application in anesthetized rats leads to hemodynamic impairment during endotoxemia, mimicking clinical cases of hemodynamic alterations during sepsis. To our knowledge, longitudinal hemodynamic measurements using a left ventricle catheterization system in rats under endotoxemic conditions in
Fig. 6 e Cardiac TNF-a mRNA expression. LPS application upregulated the expression of TNF-a compared with untreated controls. A stronger induction was observed in VGX animals in the (A) left ventricle (LV) and (B) right ventricle (RV). This effect was diminished by vagal stimulation. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5).
486
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Fig. 7 e Cardiac CrAT mRNA expression. CrAT expression was only slightly upregulated in LPS-treated animals in the (A) left ventricle (LV). In the (B) right ventricle (RV) CrAT expression was downregulated in VGX and stimulated animals. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5).
combination with a vagus stimulating treatment were performed for the first time. After vagotomy, stronger effects of depressed cardiac function are observed in endotoxemic animals. These results are also confirmed by an elevated inflammatory response, measurable by increased proinflammatory cytokine levels of IL-1b and TNF-a in the plasma and ventricular tissues. TNF-a protein level in the right ventricle is higher compared with the left ventricle after vagotomy. Analyses of TNF-a gene expression are in concordance with these results. Our results are in line with findings of an earlier study [22] showing a higher expression of TNF-a in right ventricle in left ventricular tissue after LPS stimulation. This was paradoxically associated with a less impaired function of the right ventricle. This points toward a TNF-a resistance of the right heart, with the cells being producers but not itself targets of TNF-a. Because the concept of myocardial TNF resistance is poorly understood, further studies are necessary to address this aspect regarding its molecular mechanism and the appearance in the clinical setting. Meanwhile, VNS of vagal afferents is a common therapy option in epilepsy and depression [23]. Electrical stimulation of the efferent vagal nerve fibers with low frequency is supposed to own anti-inflammatory properties and is able to activate the “cholinergic anti-inflammatory reflex” [24]. Our results approve the earlier reports of an overall beneficial effect of vagal stimulation by adding evidence about the influence on hemodynamic parameters and the anti-inflammatory quality by reducing proinflammatory cytokine levels of IL-1b and TNF-a in plasma and cardiac tissue on mRNA and protein level. To reverse both hemodynamic and immunologic effects of diminished vagal tone, even a short-term stimulation of the N. vagus seems to be sufficient during the LPS infusion. In respect to the phenomenon of autonomic dysfunction, which is commonly present during sepsis and septic shock [25], our findings add evidence for a potential use of vagus nerve stimulators to reestablish the lacking parasympathetic innervation of critical ill patients. Currently, the involvement of the spleen in the “cholinergic anti-inflammatory reflex” is discussed controversial, focusing on the anatomical absence of parasympathetic innervation of the organ [26,16,27]. Despite the anatomical
controversy, the beneficial effect of cholinergic stimulation during inflammation is indisputable and has been shown in a variety of conditions [28]. Assuming that there is no splenic attendance in this “reflex” and reconsidering the early experiments of Otto Loewi, we put the heart in the center of interest. Despite the fact, that ventricular myocardium is only sparsely innervated by the N. vagus, a strong innervation takes place at the sinoatrial node. This might be the main source of ACh released into the cardiac environment and to the circulating blood, where it could act as a basal modulator of the cardiomyocytes and immune cells. Moreover, we wanted to revisit the possibility of a nonneuronal ACh production by the cardiomyocytes themselves, as shown earlier by Kakinuma et al. [19], and especially the regulation under systemic inflammatory conditions. It was shown that the rat cardiomyocytes can express the enzymes ChAT and vesicular acetylcholine transporter necessary for ACh synthesis [29], and the importance of ACh in balancing the heart’s homeostasis during stress has been recently proven [30]. Interestingly, our data revealed no ChAT mRNA expression in the ventricular tissue, whereas the expression of CrAT, another ACh producing enzyme, could be found in left and right ventricles. These results are in line with an elderly study, in which the authors declare that ChAT activity was mainly found in the atria instead in the ventricles and ACh located in the ventricles derives mainly from CrAT [21]. The ventricular cells of the heart, which have been shown to express a variety of different ACh receptor subunits including the subunit responsible for mediating the cytokine inhibitory effect [18], could therefore be on the one hand producers of ACh themselves and on the other hand targets of it. Having in mind that myocardial cells are capable of secreting proinflammatory cytokines on sensing of circulating mediators or pathogenassociated molecular pattern, for example LPS [31], this paracrine and autocrine mechanism could be the molecular basis for our results, but further studies are needed to prove this proposed mechanisms especially regarding the secretion of ACh from the cardiac tissue. Also, experiments should be conducted with an intact N. vagus to narrow the gap between animal model and the clinical setting, as electrical stimulation might also influence “upstream” signaling events taking place
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
in the brain, and the effect of a total loss of vagal innervation especially on the vascular tone might be highly relevant in a model of hemodynamic impairment like LPS-induced septic shock.
5.
Conclusions
Our results demonstrate a beneficial impact of VNS on myocardial inflammation and hemodynamic alteration under LPS-induced systemic inflammatory conditions. We are able to show the expression of CrAT, an enzyme capable of synthesizing ACh, in the ventricular tissue with a ventriclespecific regulation under experimental conditions. In consideration of these results, we believe that the heart plays a pivotal role in the context of the “cholinergic antiinflammatory reflex”, may be as an interface between “nerve” and “body.” We also suggest that ACh derived from terminal endings of the N. vagus inside the myocard on stimulation inhibits the proinflammatory gene expression in cardiomyocytes and improves cardiac function during septic cardiomyopathy. On basis of our results, we believe that vagal stimulation has the potential to become an important therapeutic intervention in the treatment of patients with sepsis by counteracting the commonly observed autonomic dysfunction. By balancing this disturbed system guided along clinical markers like heart rate variability, a reconstitution of homeostasis might be achieved. The effect on the organism might at that point not be solely of anti-inflammatory nature; also, global changes in, for example, the vascular tone might contribute to a beneficial influence. Yet, further studies are necessary to bridge the preclinical results into a clinical application.
Acknowledgment The authors declare that they have no conflict of interest. No funding has been received for this work. Author contributions: A.S. and F.U. collected the data. A.S., C.L., M.H., M.A.W., and F.U participated in analysis and interpretation of data. A.S., C.L., M.H., M.A.W., and F.U drafted and revised the manuscript. C.L., M.A.W., and F.U. were responsible for conception and design.
references
[1] Angus DC, Wax RS. Epidemiology of sepsis: an update. Crit Care Med 2001;29(7 Suppl):S109. [2] Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29(7):1303. [3] Vincent J-L, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 2006;34(2):344.
487
[4] Engel C, Brunkhorst FM, Bone H-G, et al. Epidemiology of sepsis in Germany: results from a national prospective multicenter study. Intensive Care Med 2007;33(4):606. [5] Werdan K, Hettwer S, Bubel S, et al. Septic circulatory shock and septic cardiomyopathy. Internist (Berl) 2009;50(7):799. 802-6, 808-9. [6] Werdan K, Oelke A, Hettwer S, et al. Septic cardiomyopathy: hemodynamic quantification, occurrence, and prognostic implications. Clin Res Cardiol 2011;100(8):661. [7] Romero-Bermejo FJ, Ruiz-Bailen M, Gil-Cebrian J, HuertosRanchal MJ. Sepsis-induced cardiomyopathy. Curr Cardiol Rev 2011;7(3):163. [8] Zanotti-Cavazzoni SL, Hollenberg SM. Cardiac dysfunction in severe sepsis and septic shock. Curr Opin Crit Care 2009; 15(5):392. [9] Maass DL, White J, Horton JW. IL-1beta and IL-6 act synergistically with TNF-alpha to alter cardiac contractile function after burn trauma. Shock 2002;18(4):360. [10] Schmidt H, Muller-Werdan U, Hoffmann T, et al. Autonomic dysfunction predicts mortality in patients with multiple organ dysfunction syndrome of different age groups. Crit Care Med 2005;33(9):1994. [11] Loewi O. Uber humorale Uebertragbarkeit der Herznervenwirkung. Pflugers Arch 1924;189:239. [12] Tansey EM. Chemical neurotransmission in the autonomic nervous system: Sir Henry Dale and acetylcholine. Clin Auton Res 1991;1(1):63. [13] Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 2000;405(6785):458. [14] Tracey KJ. The inflammatory reflex. Nature 2002;420(6917): 853. [15] Rosas-Ballina M, Ochani M, Parrish WR, et al. Splenic nerve is required for cholinergic antiinflammatory pathway control of TNF in endotoxemia. Proc Natl Acad Sci 2008;105(31): 11008. [16] Bratton BO, Martelli D, McKinley MJ, Trevaks D, Anderson CR, McAllen RM. Neural regulation of inflammation: no neural connection from the vagus to splenic sympathetic neurons. Exp Physiol; 2012. [17] Kakinuma Y, Akiyama T, Okazaki K, et al. A non-neuronal cardiac cholinergic system plays a protective role in myocardium salvage during ischemic insults. PLoS ONE 2012; 7(11):e50761. [18] Dvorakova M, Lips KS, Bru¨ggmann D, Slavikova J, Kuncova J, Kummer W. Developmental changes in the expression of nicotinic acetylcholine receptor alpha-subunits in the rat heart. Cell Tissue Res 2005;319(2):201. [19] Kakinuma Y, Akiyama T, Sato T. Cholinoceptive and cholinergic properties of cardiomyocytes involving an amplification mechanism for vagal efferent effects in sparsely innervated ventricular myocardium. FEBS J 2009;276(18):5111. [20] Pacher P, Nagayama T, Mukhopadhyay P, Ba´tkai S, Kass DA. Measurement of cardiac function using pressureevolume conductance catheter technique in mice and rats. Nat Protoc 2008;3(9):1422. [21] Slavikova J, Tucek S. Choline acetyltransferase in the heart of adult rats. Pflugers Arch 1982;392(3):225. [22] Markel TA, Crisostomo PR, Wang M, Herrmann JL, Abarbanell AM, Meldrum DR. Right ventricular TNF resistance during endotoxemia: the differential effects on ventricular function. Am J Physiol Regul Integr Comp Physiol 2007;293(5):R1893. [23] Beekwilder JP, Beems T. Overview of the clinical applications of vagus nerve stimulation. J Clin Neurophysiol 2010;27(2):130. [24] Tracey KJ. Physiology and immunology of the cholinergic antiinflammatory pathway. J Clin Invest 2007;117(2):289.
488
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
[25] Ebelt H, Werdan K. Sepsis und Herz. Internist 2010;51(7):844. [26] Buijs RM, van der Vliet J, Garidou M-L, Huitinga I, Escobar C. Spleen vagal denervation inhibits the production of antibodies to circulating antigens. PLoS One 2008;3(9):e3152. [27] Bonaz B, Picq C, Sinniger V, Mayol JF, Clarenc¸on D. Vagus nerve stimulation: from epilepsy to the cholinergic antiinflammatory pathway. Neurogastroenterology Motil 2013; 25(3):208. [28] Rosas-Ballina M, Tracey KJ. Cholinergic control of inflammation. J Intern Med 2009;265(6):663.
[29] Rana OR, Schauerte P, Kluttig R, et al. Acetylcholine as an age-dependent non-neuronal source in the heart. Auton Neurosci 2010;156(1-2):82. [30] Roy A, Fields WC, Rocha-Resende C, et al. Cardiomyocytesecreted acetylcholine is required for maintenance of homeostasis in the heart. The FASEB J 2013;27. [31] Garner LB, Willis MS, Carlson DL, et al. Macrophage migration inhibitory factor is a cardiac-derived myocardial depressant factor. Am J Physiol Heart Circ Physiol 2003; 285(6):H2500.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Loss of vagal tone aggravates systemic inflammation and cardiac impairment in endotoxemic rats Astrid Schulte, MSc, Christoph Lichtenstern, MD,* Michael Henrich, DPhil, Markus A. Weigand, MD, and Florian Uhle, MSc Department of Anesthesiology and Intensive Care Medicine, Justus-Liebig-University Giessen, Giessen, Germany
article info
abstract
Article history:
Background: During the course of sepsis, often myocardial depression with hemodynamic
Received 17 September 2013
impairment occurs. Acetylcholine, the main transmitter of the parasympathetic Nervus
Received in revised form
vagus, has been shown to be of importance for the transmission of signals within the
13 January 2014
immune system and also for a variety of other functions throughout the organism. Hy-
Accepted 16 January 2014
pothesizing a potential correlation between this dysfunction and hemodynamic impair-
Available online 21 January 2014
ment, we wanted to assess the impact of vagal stimulation on myocardial inflammation and function in a rat model of lipopolysaccharide (LPS)-induced septic shock. As the
Keywords:
myocardial tissue is (sparsely) innervated by the N. vagus, there might be an important
Septic cardiomyopathy
anti-inflammatory effect in the heart, inhibiting proinflammatory gene expression in
Endotoxemia
cardiomyocytes and improving cardiac function.
Vagus nerve stimulation
Materials and methods: We performed stimulation of the right cervical branch of the N. vagus
Acetylcholine
in vagotomized, endotoxemic (1 mg/kg body weight LPS, intravenously) rats. Hemody-
Cholinergic antiinflammatory reflex
namic parameters were assessed over time using a left ventricular pressureevolume catheter. After the experiments, hearts and blood plasma were collected, and the expression of proinflammatory cytokines was measured using quantitative reverse transcription polymerase chain reaction and enzyme-linked immunosorbent assay. Results: After vagotomy, the inflammatory response was aggravated, measurable by elevated cytokine levels in plasma and ventricular tissue. In concordance, cardiac impairment during septic shock was pronounced in these animals. To reverse both hemodynamic and immunologic effects of diminished vagal tone, even a brief stimulation of the N. vagus was enough during initial LPS infusion. Conclusions: Overall, the N. vagus might play a major role in maintaining hemodynamic stability and cardiac immune homeostasis during septic shock. ª 2014 Elsevier Inc. All rights reserved.
1.
Introduction
Sepsis and septic shock are, despite intensive research and new therapeutic options, the major causes of death in
intensive care units in Western civilization [1e4]. Sepsisinduced cardiac dysfunction often occurs during the course of disease and is characterized by hemodynamic and cardiovascular instability [5e7] accompanied by increased mortality
* Corresponding author. Department of Anesthesiology and Intensive Care Medicine, Justus-Liebig-University Giessen, Rudolf-Buchheim-Straße 7, 35392 Giessen, Germany. Tel.: þ49 641 985 44401; fax: þ49 641 985 44409. E-mail address: [email protected] (C. Lichtenstern). 0022-4804/$ e see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.01.022
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
[8]. Besides occurring inflammatory processes [9], one important aspect in septic cardiomyopathy is the failure of parasympathetic innervation, leading to autonomic dysfunction with, for example, reduced variability of the heart rate [10]. At the beginning of the 20th century, Otto Loewi [11] published his results of the “Vagusstoff” experiments. He showed that the “Vagusstoff”, later identified as Acetylcholine (ACh) by Henry Dale [12], is secreted from isolated, vagus nervee stimulated frog hearts. Approximately 80 y later, it was shown that ACh, the principal vagal neurotransmitter, was able to reduce proinflammatory cytokine levels in response to endotoxin in vitro in human macrophages and in vivo through direct electrical stimulation of the efferent vagal nerve in rats [13]. These experiments highlighted the link between immune and vegetative nerve system, and the idea about the “cholinergic anti-inflammatory reflex” was postulated [14]. First, it was assumed that the spleen, innervated by catecholaminergic splenic nerve fibers originating in the celiac-superior mesenteric plexus ganglia, is required for the cytokine regulatory effect [15]. However, a recent study provided evidence that the vagal preganglionic terminals do not synapse with the sympathetic neurons innervating the spleen [16]. In this context, the question raised about a “nonneuronal cardiac cholinergic system” [17] with an influence of cardiac innervation during inflammation, introducing the heart as the possible interface between immune and nervous system. Analysis of cardiac myocytes indicated that these cells express the specific ACh receptor subunit for mediating this reflex [18] and, furthermore, are able to synthesize the enzymes necessary for producing ACh and to react in a paracrine fashion [19]. After Otto Loewi’s approach of vagal stimulating hearts, but staying in the living animal, we performed hemodynamic measurements with stimulation of the right cervical branch of Nervus vagus from Lewis rats under LPS-induced septic shock to prove our hypothesis that the N. vagusederived ACh, released from terminal endings inside the cardiac tissue on stimulation, inhibits the proinflammatory gene expression in cardiomyocytes and improves cardiac function during septic cardiomyopathy.
2.
Material and methods
2.1.
Animal handling and care
Male Lewis rats (275e300 g) were purchased from Charles River (Sulzfeld, Germany). Animals were kept in cages with free access to water and food on a 12-h light and dark cycle. All procedures involving animals conform to the standards for animal experiments and were approved by the local committee for animal care (No. 108-2010, Regierungsprasidium Giessen, Germany).
2.2.
Animal model of endotoxic shock
Experiments were performed in anesthetized rats. Anesthesia was induced (5%) and maintained (2%e3%) with isoflurane (Baxter, Unterschleissheim, Germany). After endotracheal intubation with a 16-G catheter, animals were ventilated with a rodent respirator (Harvard Inspira, MA) in a weight-adjusted
481
manner. Electrocardiogram electrodes were placed, and the heart rate was recorded. Body temperature was rectally monitored and maintained around 37 C by a heating pad (Harvard Apparatus, MA) and a red light heating lamp. Ringer solution (Braun, Melsungen, Germany; 10 mL/kg/h) and Fentanyl (Ratiopharm, Ulm, Germany; 10 mg/kg/h) were injected intravenously over the lateral tail vein through a syringe pump (Braun). A Micro-Tip catheter (SPR-1000; Millar Instruments, Houston, TX) was inserted in the tail artery for continuous measurement of the arterial blood pressure. Animals were divided in a right cervical vagotomy group with no stimulation (vagotomized [VGX] þ LPS, n ¼ 5), short electrical stimulation (VGX þ LPS þ 2 min stim, n ¼ 6), long electrical stimulation (VGX þ LPS þ 20 min stim, n ¼ 6), and two groups (sham, n ¼ 6 and sham þ LPS, n ¼ 6) without vagotomy. The right trunk of the N. vagus was gently exposed in all animals, but only transected in the VGX rats. For electrical stimulation in the VGX animals, the vagal nerve trunk was placed between the electrodes that were connected to a constant voltage stimulator (DeMeTec, Langgoens, Germany). Stimulation (5 V, 2 ms, and 1 Hz) was performed for 20 min, starting 5 min before LPS injection (t0) or for 2 min (1 min before and 1 min after LPS administration). As a result of its amphiphile structure, LPS (Escherichia coli 0111:B4, LPS-EB Ultrapure; InvivoGen, San Diego, CA; 5 mg/mL in saline) is capable of forming micelles in aqueous solution. Therefore, it was sonicated for 30 min before injection. All animals, except the sham group, received an LPS bolus (1 mg/kg in 1 mL saline, intravenously over tail vein) more than 20 min via a syringe pump (Harvard Apparatus). Experiments were finished when the animals died during the experimental protocol or 6 h after LPS or placebo (0.9 % saline) administration. After euthanasia, blood samples were collected and centrifuged, and the plasma was stored at 80 C. Hearts were excised, dissected into left and right ventricle, and flash frozen in liquid nitrogen. Storage was carried out at 80 C.
2.3.
Hemodynamic measurements
Cardiac function studies of the left ventricle were performed inserting a pressureevolume conductance catheter (SPR-838; Millar Instruments, Houston, TX), using the closed chest approach [20]. Under deep anesthesia, the right carotid artery was exposed and separated from surrounding connective tissue and the vagus nerve. The artery was punctured with a 26-G needle, and the catheter was inserted and fixed with a suture and carefully moved into the left ventricle. Preparation and catheterization were accomplished in approximately 1 h, followed by a recovery time of 30 min before the LPS bolus. Pressureevolume signals were digitized with a PowerLab 8/30 signal convertor (ADInstruments, Spechbach, Germany) and recorded using the LabChart7 Pro Software for Windows (ADInstruments). In each individual animal, parallel conductance catheter calibration was performed; therefore, 30 mL of hypertonic saline (10%) was injected through a polyethylene catheter into the right jugular vein. Once, cuvette calibration for blood conductivity determination was performed with fresh heparinized, warm blood. The evaluation of the recorded data was performed with PVAN 1.1 (Millar Instruments), and hemodynamic parameters (cardiac output [CO], ejection fraction [EF], and heartbeat) were determined.
482
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Fig. 1 e Hemodynamic measurements in control and endotoxemic rats. Heart rate (HR), (A) and blood pressure (B) decreased during 6-h anesthesia in LPS-treated rats and in control animals. LPS also impaired CO, (C) and EF, (D). Animals at risk (E). Controls are depicted by dotted lines; LPS animals are displayed by solid lines. n [ 6 at t0. Data are mean ± standard error of mean.
2.4.
Protein isolation from heart tissue
Frozen heart tissue (50e100 mg) was homogenized with a bead mill (Schwingmuehle NM 301; Retsch, Haan, Germany) in 1 mL 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid buffer (20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 1.5 mM ethylenediaminetetraacetate, pH 7.4, Protease Inhibitor [Roche, Mannheim, Germany]). After centrifugation, supernatant was collected, and total protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific, Rockford, IL).
2.5.
Cytokine enzyme-linked immunosorbent assay
Plasma samples and proteins from left and right ventricle were centrifuged, and enzyme-linked immunosorbent assay was performed using Quantikine Rat TNF-a (RTA00) and interleukin 1b (IL-1b; RLB00) kits (R&D Systems, Abingdon, UK) according to the manufacturer’s information. In case of tissue sample, cytokine levels were normalized against total protein content.
2.6. RNA Isolation, complementary DNA synthesis, and quantitative real-time polymerase chain reaction For RNA extraction, the right and left ventricles (50e100 mg) were homogenized with a bead mill (Schwingmuehle NM 301; Retsch) in 1 mL TRIzol Reagent (Ambion, Carlsbad, CA)
and then stored at room temperature for 5 min. After centrifugation, the supernatant was removed in a new tube and 0.2 mL chloroform (Roth, Karlsruhe, Germany) was added to the homogenate, mixed and allowed to stand at room temperature for 2 min. After incubation, the sample was centrifuged again at 12.000g for 15 min at 4 C. The aqueous phase was carefully removed and transferred to a new tube. An equal volume of 70% ethanol was added. Afterward, the sample was loaded on an RNeasy Mini Kit column (Qiagen, Hilden, Germany). At this step, isolation was performed according to the manufacturer’s protocol, and RNA was stored at 80 C. For analysis, 0.5 mg of RNA was reversely transcribed using the QuantiTect Reverse Transcription Kit (Qiagen). Quantitative real-time polymerase chain reaction (qRt-PCR) was performed using TaqMan Gene Expression Master Mix (Applied Biosystems, Foster City, CA). Assay numbers were as follows: Rn01453442_m1 (choline acetyltransferase [ChAT]), Rn01758585_m1 (carnitine acetyltransferase [CrAT]), Rn01525859_g1 (tumor necrosis factor alpha [TNF-a]) and Rn00667869_m1 (b-Actin). For data analysis, the comparative DCt method was used (2^[Ct ActinCt gen-of-interest]).
2.7.
Statistical analysis
All statistical analysis was performed using Graphpad Prism version 5.0f for Mac (GraphPad Software, La Jolla, CA). Group comparisons were done using the KruskaleWallis test followed by Dunn’s post-test. KaplaneMeier graphs were drawn
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
483
(369 13.5 bpm) after LPS bolus in comparison with untreated animals (353 25.8 bpm) but reaches the same level again at time point 3.5 h (LPS: 336 17.8 bpm versus sham: 335 30.7 bpm; Fig. 1A). Depression of arterial blood pressure was more distinctive in the LPS animals as in the untreated controls (Fig. 1B), which pointed to an LPS-induced effect. CO and EF were both increased after LPS application (Fig. 1C and D) but showed a strong reduction and impairment of cardiac function after 3 h of LPS administration.
3.3.
Survival time in endotoxemic, VGX rats
Survival time of VGX animals in combination with LPS injection was reduced compared with endotoxemic animals with an intact vagus nerve (Fig. 2B in comparison with Fig. 2A). All rats (n ¼ 5) from the VGX þ LPS group died within 2 h after LPS application, two died after 1 h and accordingly after 1.5 h.
3.4. Effects of electrical stimulation of the N. vagus in VGX, endotoxemic rats
Fig. 2 e Survival analysis of endotoxemic, VGX, and stimulated rats. KaplaneMeier curve of control and LPStreated animals (A). No mortality occurred in the control group, only one animal remained alive after 6 h following LPS-injection. KaplaneMeier curve of endotoxemic, VGX, and stimulated rats (B). Survival time was reduced because of vagotomy and LPS application but was improved by VNS (n ‡ 5). for visualization of survival time, and the groups were compared using the log-rank test.
3.
To determine whether the loss of vagal tone could be reestablished by constant voltage pulses, animals underwent vagotomy with vagus nerve stimulation (VNS) for 2 min (VGX þ LPS þ stim_2 min, n ¼ 6) or 20 min (VGX þ LPS þ stim_20 min, n ¼ 6). Electrical stimulation improved survival time (Fig. 2B) in VGX, LPS-induced septic shock rats compared with unstimulated, VGX controls. Heart rate was only slightly decreased; animals with short-term stimulation showed even a small increase (Fig. 3A). Arterial blood pressure was strongly reduced in unstimulated, VGX rats after LPS administration, whereas VNS effected an attenuate reduction (Fig. 3B). Hemodynamic parameters (CO, EF) also showed less impairment (Fig. 3C and D) with stimulation after LPS application, but after about 3 h, a strong decrease was observed as well.
Results
3.1. Hemodynamic measurements in rats more than 6 h under anesthesia For our prolonged hemodynamic experiments, we chose the closed chest approach for left ventricle catheterization, because of its less invasive character [20]. During a 6-h anesthesia period with isoflurane in catheterized, untreated Lewis rats (n ¼ 6), a decreased heart rate (Fig. 1A) and arterial blood pressure (Fig. 1B) could be observed over the experimental period. CO (Fig. 1C) and EF (Fig. 1D) remained constant over the whole duration of the experiments. For optimal maintenance of the vital functions and to stabilize the animals for a longer experimental period, body temperature was kept at about 37 C, and an adequate amount of fluid was substituted.
3.2. Hemodynamic measurements in rats after endotoxemia Among the LPS group (n ¼ 6), only one animal survived the whole duration of the experiment, one died after 5 h, three after 4.5 h and another one 4 h after LPS injection (Fig. 2A). Heart rate in LPS-treated rats was slightly increased at 1.5 h
3.5. Cytokine TNF-a and IL-1b levels in plasma and cardiac tissue For evaluation of the impact of vagal stimulation, cytokine levels of TNF-a and IL-1b in plasma and left and right ventricular tissues were determined using enzyme-linked immunosorbent assay. In plasma, LPS application led to a slightly increase of TNF-a and IL-1b compared with the untreated control group (Fig. 4A and B). In VGX rats, TNF-a (P < 0.001) and IL-1b levels were markedly elevated, whereas VNS moderated this increase (Fig. 4A and B). Expression patterns of TNF-a in the left and right ventricles corresponded to the patterns that were found in the plasma (Fig. 5A and B). The inflammatory response was increased by vagotomy, whereas VNS attenuated this effect. Short-term stimulation showed a trend to be more effective, but this was not statistically significant. Noteworthy, IL-1b levels in the left and right ventricles were elevated in animals receiving LPS and VGX þ LPS but were further increased in stimulated animals (Fig. 5C and D). Contrary, the highest levels of IL-1b were measured by shortterm stimulation (P < 0.01, compared with untreated controls). Remarkably, cytokine patterns of left and right ventricles differed in expression (Fig. 5).
484
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Fig. 3 e Effects of electrical VNS in VGX, endotoxemic rats. (A) Heart rate (HR), (B) blood pressure (C) CO, and (D) EF are strongly reduced in VGX, endotoxemic animals. Hemodynamic parameters were ameliorated in the vagal stimulation groups in comparison with nonstimulated rats. Animals at risk (E). Controls (VGX) are the dotted line, animals with short stimulation (VGX D 2 min) are displayed by triangle line, and animals with long stimulation (VGX D 20 min) are displayed by square line. n ‡ 5 at t0. Data are mean ± standard error of mean.
3.6. tissue
TNF-a messenger RNA expression in ventricular
To further confirm the results of inflammatory response on gene expression level in the right and left ventricles, qRT-PCR was performed. TNF-a messenger RNA (mRNA) expression was upregulated by the induction of LPS compared with untreated controls. Levels in VGX animals were significantly higher (P < 0.001 in the right ventricle). Consistent with the findings on protein level, TNF-a mRNA expression in the
stimulation groups was on a lower level. The lowest expression was found in animals with short-term stimulation, but it remained above the level measured in control animals (Fig. 6A and B).
3.7. tissue
ChAT and CrAT mRNA expression in ventricular
To determine if myocardial cells are able to produce ACh, qRTPCR was performed against ChAT, the ACh producing enzyme.
Fig. 4 e Plasma cytokine protein levels. Elevated proinflammatory TNF-a (A) and IL-1b (B) levels were measured in VGX, endotoxemic animals. Vagal stimulation decreased the plasma levels of both mediators, however, they remained above the levels of controls. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5).
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
485
Fig. 5 e Cardiac cytokine protein levels. TNF-a was increased in endotoxemic, VGX rats in the (A) left ventricle (LV) and (B) right ventricle (RV). On VNS, it was less enhanced. Levels of IL-1b protein were slightly elevated in the LPS and LPS D VGX group but to a greater extent in the stimulation groups both in left (C) and right (D) ventricular tissue. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5, except left ventricle IL-1b: n ‡ 3).
Mentionable, mRNA expression of ChAT was not detectable in any ventricular tissue of the animals. ACh can also be synthesized by CrAT. Slavikova et al. [21] have shown that ChAT activity is very low in the rat ventricles. They suspected that synthesized ACh in the ventricles is a product of CrAT. This former observation could be confirmed by qRT-PCR analysis in this study. CrAT mRNA expression was detectable in all groups with no significant regulation in the left ventricle (Fig. 7). In the right ventricle, downregulation of CrAT was observed in LPS-treated animals with the strongest reduction in VGX rats (P < 0.01, Fig. 7B). Eventually, ACh in the ventricles is synthesized by CrAT, and this enzyme is downregulated
significantly in VGX rats on LPS stimulation in the right ventricle.
4.
Discussion
The present study provides evidence that LPS application in anesthetized rats leads to hemodynamic impairment during endotoxemia, mimicking clinical cases of hemodynamic alterations during sepsis. To our knowledge, longitudinal hemodynamic measurements using a left ventricle catheterization system in rats under endotoxemic conditions in
Fig. 6 e Cardiac TNF-a mRNA expression. LPS application upregulated the expression of TNF-a compared with untreated controls. A stronger induction was observed in VGX animals in the (A) left ventricle (LV) and (B) right ventricle (RV). This effect was diminished by vagal stimulation. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5).
486
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
Fig. 7 e Cardiac CrAT mRNA expression. CrAT expression was only slightly upregulated in LPS-treated animals in the (A) left ventricle (LV). In the (B) right ventricle (RV) CrAT expression was downregulated in VGX and stimulated animals. Data are mean ± standard error of mean (*P < 0.05, **P < 0.01, and ***P < 0.001, n ‡ 5).
combination with a vagus stimulating treatment were performed for the first time. After vagotomy, stronger effects of depressed cardiac function are observed in endotoxemic animals. These results are also confirmed by an elevated inflammatory response, measurable by increased proinflammatory cytokine levels of IL-1b and TNF-a in the plasma and ventricular tissues. TNF-a protein level in the right ventricle is higher compared with the left ventricle after vagotomy. Analyses of TNF-a gene expression are in concordance with these results. Our results are in line with findings of an earlier study [22] showing a higher expression of TNF-a in right ventricle in left ventricular tissue after LPS stimulation. This was paradoxically associated with a less impaired function of the right ventricle. This points toward a TNF-a resistance of the right heart, with the cells being producers but not itself targets of TNF-a. Because the concept of myocardial TNF resistance is poorly understood, further studies are necessary to address this aspect regarding its molecular mechanism and the appearance in the clinical setting. Meanwhile, VNS of vagal afferents is a common therapy option in epilepsy and depression [23]. Electrical stimulation of the efferent vagal nerve fibers with low frequency is supposed to own anti-inflammatory properties and is able to activate the “cholinergic anti-inflammatory reflex” [24]. Our results approve the earlier reports of an overall beneficial effect of vagal stimulation by adding evidence about the influence on hemodynamic parameters and the anti-inflammatory quality by reducing proinflammatory cytokine levels of IL-1b and TNF-a in plasma and cardiac tissue on mRNA and protein level. To reverse both hemodynamic and immunologic effects of diminished vagal tone, even a short-term stimulation of the N. vagus seems to be sufficient during the LPS infusion. In respect to the phenomenon of autonomic dysfunction, which is commonly present during sepsis and septic shock [25], our findings add evidence for a potential use of vagus nerve stimulators to reestablish the lacking parasympathetic innervation of critical ill patients. Currently, the involvement of the spleen in the “cholinergic anti-inflammatory reflex” is discussed controversial, focusing on the anatomical absence of parasympathetic innervation of the organ [26,16,27]. Despite the anatomical
controversy, the beneficial effect of cholinergic stimulation during inflammation is indisputable and has been shown in a variety of conditions [28]. Assuming that there is no splenic attendance in this “reflex” and reconsidering the early experiments of Otto Loewi, we put the heart in the center of interest. Despite the fact, that ventricular myocardium is only sparsely innervated by the N. vagus, a strong innervation takes place at the sinoatrial node. This might be the main source of ACh released into the cardiac environment and to the circulating blood, where it could act as a basal modulator of the cardiomyocytes and immune cells. Moreover, we wanted to revisit the possibility of a nonneuronal ACh production by the cardiomyocytes themselves, as shown earlier by Kakinuma et al. [19], and especially the regulation under systemic inflammatory conditions. It was shown that the rat cardiomyocytes can express the enzymes ChAT and vesicular acetylcholine transporter necessary for ACh synthesis [29], and the importance of ACh in balancing the heart’s homeostasis during stress has been recently proven [30]. Interestingly, our data revealed no ChAT mRNA expression in the ventricular tissue, whereas the expression of CrAT, another ACh producing enzyme, could be found in left and right ventricles. These results are in line with an elderly study, in which the authors declare that ChAT activity was mainly found in the atria instead in the ventricles and ACh located in the ventricles derives mainly from CrAT [21]. The ventricular cells of the heart, which have been shown to express a variety of different ACh receptor subunits including the subunit responsible for mediating the cytokine inhibitory effect [18], could therefore be on the one hand producers of ACh themselves and on the other hand targets of it. Having in mind that myocardial cells are capable of secreting proinflammatory cytokines on sensing of circulating mediators or pathogenassociated molecular pattern, for example LPS [31], this paracrine and autocrine mechanism could be the molecular basis for our results, but further studies are needed to prove this proposed mechanisms especially regarding the secretion of ACh from the cardiac tissue. Also, experiments should be conducted with an intact N. vagus to narrow the gap between animal model and the clinical setting, as electrical stimulation might also influence “upstream” signaling events taking place
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
in the brain, and the effect of a total loss of vagal innervation especially on the vascular tone might be highly relevant in a model of hemodynamic impairment like LPS-induced septic shock.
5.
Conclusions
Our results demonstrate a beneficial impact of VNS on myocardial inflammation and hemodynamic alteration under LPS-induced systemic inflammatory conditions. We are able to show the expression of CrAT, an enzyme capable of synthesizing ACh, in the ventricular tissue with a ventriclespecific regulation under experimental conditions. In consideration of these results, we believe that the heart plays a pivotal role in the context of the “cholinergic antiinflammatory reflex”, may be as an interface between “nerve” and “body.” We also suggest that ACh derived from terminal endings of the N. vagus inside the myocard on stimulation inhibits the proinflammatory gene expression in cardiomyocytes and improves cardiac function during septic cardiomyopathy. On basis of our results, we believe that vagal stimulation has the potential to become an important therapeutic intervention in the treatment of patients with sepsis by counteracting the commonly observed autonomic dysfunction. By balancing this disturbed system guided along clinical markers like heart rate variability, a reconstitution of homeostasis might be achieved. The effect on the organism might at that point not be solely of anti-inflammatory nature; also, global changes in, for example, the vascular tone might contribute to a beneficial influence. Yet, further studies are necessary to bridge the preclinical results into a clinical application.
Acknowledgment The authors declare that they have no conflict of interest. No funding has been received for this work. Author contributions: A.S. and F.U. collected the data. A.S., C.L., M.H., M.A.W., and F.U participated in analysis and interpretation of data. A.S., C.L., M.H., M.A.W., and F.U drafted and revised the manuscript. C.L., M.A.W., and F.U. were responsible for conception and design.
references
[1] Angus DC, Wax RS. Epidemiology of sepsis: an update. Crit Care Med 2001;29(7 Suppl):S109. [2] Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29(7):1303. [3] Vincent J-L, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 2006;34(2):344.
487
[4] Engel C, Brunkhorst FM, Bone H-G, et al. Epidemiology of sepsis in Germany: results from a national prospective multicenter study. Intensive Care Med 2007;33(4):606. [5] Werdan K, Hettwer S, Bubel S, et al. Septic circulatory shock and septic cardiomyopathy. Internist (Berl) 2009;50(7):799. 802-6, 808-9. [6] Werdan K, Oelke A, Hettwer S, et al. Septic cardiomyopathy: hemodynamic quantification, occurrence, and prognostic implications. Clin Res Cardiol 2011;100(8):661. [7] Romero-Bermejo FJ, Ruiz-Bailen M, Gil-Cebrian J, HuertosRanchal MJ. Sepsis-induced cardiomyopathy. Curr Cardiol Rev 2011;7(3):163. [8] Zanotti-Cavazzoni SL, Hollenberg SM. Cardiac dysfunction in severe sepsis and septic shock. Curr Opin Crit Care 2009; 15(5):392. [9] Maass DL, White J, Horton JW. IL-1beta and IL-6 act synergistically with TNF-alpha to alter cardiac contractile function after burn trauma. Shock 2002;18(4):360. [10] Schmidt H, Muller-Werdan U, Hoffmann T, et al. Autonomic dysfunction predicts mortality in patients with multiple organ dysfunction syndrome of different age groups. Crit Care Med 2005;33(9):1994. [11] Loewi O. Uber humorale Uebertragbarkeit der Herznervenwirkung. Pflugers Arch 1924;189:239. [12] Tansey EM. Chemical neurotransmission in the autonomic nervous system: Sir Henry Dale and acetylcholine. Clin Auton Res 1991;1(1):63. [13] Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 2000;405(6785):458. [14] Tracey KJ. The inflammatory reflex. Nature 2002;420(6917): 853. [15] Rosas-Ballina M, Ochani M, Parrish WR, et al. Splenic nerve is required for cholinergic antiinflammatory pathway control of TNF in endotoxemia. Proc Natl Acad Sci 2008;105(31): 11008. [16] Bratton BO, Martelli D, McKinley MJ, Trevaks D, Anderson CR, McAllen RM. Neural regulation of inflammation: no neural connection from the vagus to splenic sympathetic neurons. Exp Physiol; 2012. [17] Kakinuma Y, Akiyama T, Okazaki K, et al. A non-neuronal cardiac cholinergic system plays a protective role in myocardium salvage during ischemic insults. PLoS ONE 2012; 7(11):e50761. [18] Dvorakova M, Lips KS, Bru¨ggmann D, Slavikova J, Kuncova J, Kummer W. Developmental changes in the expression of nicotinic acetylcholine receptor alpha-subunits in the rat heart. Cell Tissue Res 2005;319(2):201. [19] Kakinuma Y, Akiyama T, Sato T. Cholinoceptive and cholinergic properties of cardiomyocytes involving an amplification mechanism for vagal efferent effects in sparsely innervated ventricular myocardium. FEBS J 2009;276(18):5111. [20] Pacher P, Nagayama T, Mukhopadhyay P, Ba´tkai S, Kass DA. Measurement of cardiac function using pressureevolume conductance catheter technique in mice and rats. Nat Protoc 2008;3(9):1422. [21] Slavikova J, Tucek S. Choline acetyltransferase in the heart of adult rats. Pflugers Arch 1982;392(3):225. [22] Markel TA, Crisostomo PR, Wang M, Herrmann JL, Abarbanell AM, Meldrum DR. Right ventricular TNF resistance during endotoxemia: the differential effects on ventricular function. Am J Physiol Regul Integr Comp Physiol 2007;293(5):R1893. [23] Beekwilder JP, Beems T. Overview of the clinical applications of vagus nerve stimulation. J Clin Neurophysiol 2010;27(2):130. [24] Tracey KJ. Physiology and immunology of the cholinergic antiinflammatory pathway. J Clin Invest 2007;117(2):289.
488
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 8 ( 2 0 1 4 ) 4 8 0 e4 8 8
[25] Ebelt H, Werdan K. Sepsis und Herz. Internist 2010;51(7):844. [26] Buijs RM, van der Vliet J, Garidou M-L, Huitinga I, Escobar C. Spleen vagal denervation inhibits the production of antibodies to circulating antigens. PLoS One 2008;3(9):e3152. [27] Bonaz B, Picq C, Sinniger V, Mayol JF, Clarenc¸on D. Vagus nerve stimulation: from epilepsy to the cholinergic antiinflammatory pathway. Neurogastroenterology Motil 2013; 25(3):208. [28] Rosas-Ballina M, Tracey KJ. Cholinergic control of inflammation. J Intern Med 2009;265(6):663.
[29] Rana OR, Schauerte P, Kluttig R, et al. Acetylcholine as an age-dependent non-neuronal source in the heart. Auton Neurosci 2010;156(1-2):82. [30] Roy A, Fields WC, Rocha-Resende C, et al. Cardiomyocytesecreted acetylcholine is required for maintenance of homeostasis in the heart. The FASEB J 2013;27. [31] Garner LB, Willis MS, Carlson DL, et al. Macrophage migration inhibitory factor is a cardiac-derived myocardial depressant factor. Am J Physiol Heart Circ Physiol 2003; 285(6):H2500.
