Electrical Vagus Nerve Stimulation Attenuates Systemic Inflammation and Improves Survival in a Rat Heatstroke Model

This study was performed to gain insights into novel therapeutic approaches for the treatment of heatstroke. The central nervous system regulates peripheral immune responses via the vagus nerve, the primary neural component of the cholinergic anti-inflammatory pathway. Electrical vagus nerve stimulation (VNS) reportedly suppresses pro-inflammatory cytokine release in several models of inflammatory disease. Here, we evaluated whether electrical VNS attenuates severe heatstroke, which induces a systemic inflammatory response. Anesthetized rats were subjected to heat stress (41.5°C for 30 minutes) with/without electrical VNS. In the VNS-treated group, the cervical vagus nerve was stimulated with constant voltage (10 V, 2 ms, 5 Hz) for 20 minutes immediately after completion of heat stress. Sham-operated animals underwent the same procedure without stimulation under a normothermic condition. Seven-day mortality improved significantly in the VNS-treated group versus control group. Electrical VNS significantly suppressed induction of pro-inflammatory cytokines such as tumor necrosis factor-α and interleukin-6 in the serum 6 hours after heat stress. Simultaneously, the increase of soluble thrombomodulin and E-selectin following heat stress was also suppressed by VNS treatment, suggesting its protective effect on endothelium. Immunohistochemical analysis using tissue preparations obtained 6 hours after heat stress revealed that VNS treatment attenuated infiltration of inflammatory (CD11b-positive) cells in lung and spleen. Interestingly, most cells with increased CD11b positivity in response to heat stress did not express α7 nicotinic acetylcholine receptor in the spleen. These data indicate that electrical VNS modulated cholinergic anti-inflammatory pathway abnormalities induced by heat stress, and this protective effect was associated with improved mortality. These findings may provide a novel therapeutic strategy to combat severe heatstroke in the critical care setting.


Introduction
During the summer season, heat waves are responsible for a large number of deaths in various parts of the world. Particularly in Japan, there is a growing concern in regard to prevention and treatment of heatstroke because of restrictions placed on power consumption in the aftermath of the Fukushima nuclear disaster, which followed the Great East Japan Earthquake on March 11, 2011 [1]. Heatstroke is a life-threatening disease characterized by hyperthermia associated with a systemic inflammatory response leading to multiple organ dysfunction or failure, including hemorrhage and necrosis in the brain, lungs, heart, gastrointestinal tract, liver, and kidneys [2,3]. Although various strategies for the treatment of heatstroke-induced inflammation have been used clinically over the past several decades, the clinical efficacy of therapeutic interventions has been limited. Mortality rate in heatstroke patients over the last 50 years has remained in the 10% to 50% range [4]. Thus, a novel therapeutic strategy other than supportive care is required to improve patient outcome.
Previously, in 2000, Borovikova et al. discovered that the central nervous system (CNS), through the vagus nerve, can modulate the level of circulating tumor necrosis factor (TNF)-a induced by endotoxin [5]. This newly identified mechanism, termed the cholinergic anti-inflammatory pathway, is based on the release of acetylcholine, the principle neurotransmitter of the vagus nerve that inhibits the production of proinflammatory cytokines via its a7 nicotinic acetylcholine receptor (a7nAChR) in resident tissue macrophages [6]. Electrical or pharmacological stimulation of the vagus nerve can modulate immune responses in life-threatening conditions such as hemorrhagic shock [7,8], ischemia and reperfusion injury [9,10], or sepsis [11][12][13].
CNS disorders are predominant in heatstroke patients. Cerebral dysfunctions that occur during heatstroke include delirium, convulsion, and coma due to cerebral edema and ischemia [14,15]. In a rodent model of heatstroke, cellular damage markers are upregulated in the hypothalamus, the essential thermoregulatory center in the brain [16,17]. Thus, we hypothesized that the vagus nerve will also be affected by heat stress and that the cholinergic anti-inflammatory pathway might play a pivotal role in the pathogenesis of heatstroke, and therefore, we investigated the effect of electrical vagus nerve stimulation (VNS) in a rodent model of heatstroke.

Animals
Specific pathogen-free male Wistar rats weighing 250 to 300 g were obtained from Nihon SLC (Hamamatsu, Japan) and were allowed free access to food and water. All animal experiments were conducted in accordance with guidelines of the Animal Care and Use Committee of Osaka University Graduate School of Medicine and were approved by that committee. Rats were anesthetized by intraperitoneal injection of sodium pentobarbital (50 mg/kg body weight) (Dainippon Sumitomo Pharma Co., Osaka, Japan).

Induction of heat stress
Rat rectal temperatures were monitored continuously throughout the induction of heat stress. Anesthetized rats were subjected to environmental heat stress using a temperature control device with a warm blanket (BWT-100A; Bio Research Center, Nagoya, Japan).
The time schedule in this experimental study is shown in Figure 1. The murine model of heatstroke used in this study has been detailed previously [18][19][20]. Before induction of heat stress, the core temperature of the anesthetized rats was maintained at about 36uC with a heating pad. Next, heat stress was induced by increasing the rectal temperature approximately 1uC every 5 minutes. From the instant the rectal temperature reached 41.5uC, the core temperature was maintained at 41.5 6 0.2uC for 30 minutes. Then, the heating pad was removed, and the rats were allowed to recover at room temperature (24uC). Finally, saline was injected subcutaneously at a dose of 35 ml/kg to prevent dehydration. The time at which rats were removed from the temperature control device was defined as 0 hours.

Vagus nerve stimulation
Before induction of the heat stress, a small incision was made to expose the left cervical vagus nerve, and a platinum electrode was then placed under the exposed nerve. Immediately after the completion of heat stress, the platinum electrode was attached to the stimulation device (Master-8 Pulse Stimulator; A.M.P.I., Jerusalem, Israel). Electrical VNS was applied for 20 min at 10 V, 2 ms, and 5 Hz. Following termination of VNS, the electrode was removed, and the incision was closed with interrupted silk sutures.

Experimental design
Animals were randomly assigned to the sham group, heatstroke without stimulation group (HS-cont group), or VNS-treated heatstroke group (HS-VNS group). The sham group underwent the same experimental procedure but without application of heat stress. Immediately after heat stress, rats in the HS-VNS group received electrical VNS. In the sham group and the HS-cont group, the stimulation electrode was placed immediately under the vagus nerve, but the nerve was not stimulated.
The rats were observed over the 7 days after heat stress in order to calculate survival rates (n = 23 each in the HS-cont and HS-VNS groups and n = 8 in the sham group). Separate animals were used for blood sampling and histological analysis. Blood samples were collected at 1, 3, and 6 hours after heat stress (n = 10 for each time point in the HS-cont and HS-VNS groups and n = 5 for each time point in the sham group). Perfusion fixation was carried out to prepare tissue specimens at 6 hours after heat stress (n = 3 each in the sham, HS-cont, and HS-VNS groups, respectively).

Measurement of serum levels of TNF-a, interleukin-6, soluble thrombomodulin, and soluble E-selectin
The rats were re-anesthetized by intraperitoneal injection of pentobarbital sodium at the allocated time and underwent laparotomy, after which the abdominal aorta was cannulated with a PE50 polypropylene tube (Becton Dickinson Co, Franklin Lakes, NJ). Blood samples were collected through this tube, and the serum was isolated by centrifugation at 3000 x g for 15 minutes and frozen at -30uC until measurement.
The serum concentrations of TNF-a, interleukin-6 (IL-6), soluble thrombomodulin (sTM), and soluble E-selectin (sE-selectin) were measured with commercially available ELISA kits (TNF-a and IL-6: R&D Systems, Minneapolis, MN; sTM and E-selectin: Cusabio Biotech, Wuhan, China). Histological examinations Tissue fixation.. At 6 hours after the completion of heat stress, the rats were anesthetized by intraperitoneal injection of pentobarbital sodium and then perfused transcardially with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB). The lung and spleen were dissected, immersed in the same fixative at 4uC for 6 hours, and cryoprotected in a series of sucrose solutions (15%, 20%, and 25% sucrose in 0.1 M PB) at 4uC for 3 days. After the specimens were frozen in OCT compound (Sakura Finetechnical Co. Ltd., Osaka, Japan), they were sliced into 7-mm-thick sections by cryostat (CM3050S; Leica Microsystems, Wetzlar, Germany), and the sections were mounted on slides for staining.

Statistical analysis
Data in the figures are expressed as group means 6 standard error of the mean and were analyzed by analysis of variance followed by post-hoc Dunnett multiple comparison test. Survival curves were calculated by the Kaplan-Meier method and compared by log-rank test. A p value of ,0.05 was considered to indicate statistical significance. All statistical analyses were performed with IBM SPSS Statistics version 19.0 for Windows (SPSS Inc., Chicago, IL).

Survival assay
Rat survival curves are shown in Figure 2. The survival rate at 7 days after heat stress in the HS-cont group was 26% (6 of 23 rats), whereas that in the HS-VNS group was 61% (14 of 23 rats). When compared with the survival rate of heatstroke rats without treatment, VNS-treated heatstroke rats had significantly higher values of percent survival (p = 0.016 by log-rank test).

Serum levels of TNF-a, IL-6, sTM, and sE-selectin after heat stress
The HS-cont group showed significantly higher levels of TNF-a and IL-6 compared with those in the sham animals at every time point. In the HS-cont group, the mean value of the serum TNF-a level peaked at 3 hours after heat stress and then decreased. Electrical VNS significantly decreased the induction of serum TNF-a level at 6 hours after heat stress (p ,0.05) ( Figure 3A). Serum levels of IL-6 in the HS-cont group gradually increased after heat stress and peaked at 6 hours. These elevations were significantly suppressed at 3 and 6 hours after heat stress by VNS treatment (p ,0.05) ( Figure 3B).
To evaluate the effects of VNS treatment on endothelium in heat stress, we assessed the serum levels of sTM and E-selectin, the markers for endothelial injury. These markers were significantly higher in the HS-cont group compared with those in the sham group at 3 and 6 hours after heat stress. In the HS-VNS group, the serum level of sTM at 3 and 6 hours was significantly suppressed in comparison with that in the HS-cont group (p ,0.05) ( Figure  3C). Similarly, VNS treatment induced a significant decrease in serum level of sE-selectin at 3 hours after heat stress (p ,0.05) ( Figure 3D).

Immunohistochemical detection of a7nAChR expression in CD11b-positive cells in the spleen
To study the inflammatory cell surface expression of a7nAChR immunohistochemically, spleen sections obtained 6 hours after the heat stress were double stained with anti-rat CD11b (a surface marker for inflammatory cells) and anti-a7nAChR antibodies ( Figure 4). The number of CD11b-positive cells was upregulated in the HS-cont group compared with that in the sham group and was suppressed in the HS-VNS group compared with that in the HS-cont group. Interestingly, we found that most of the CD11b-

Immunohistochemical detection of a7nAChR expression in CD11b-positive cells and endothelial cells in the lung
To investigate the infiltration of inflammatory cells along with their expression of a7nAChR in heat stress-induced lung injury, lung sections were obtained 6 hours after the heat stress ( Figure  5). The number of CD11b-positive cells was upregulated in the HS-cont group compared with that in the sham group and was suppressed in the HS-VNS group compared with that in the HScont group. Unlike in the spleen, most of the CD11b-positive cells induced by heat stress expressed a7nAChR in the lung.
Next, we estimated alteration in the endothelial expression of a7nAChR. Immunohistochemical double staining was performed with antibodies for a7nAChR and RECA-1, which specifically recognizes rat endothelial cells. Staining revealed that the expression of a7nAChR in the lung endothelial cells also tended to be increased by heat stress.

Discussion
Heatstroke syndrome is comprised of a wide range of thermoregulatory, inflammatory, coagulatory, immune, and tissue injury responses, and presently, understanding of the endogenous mechanisms is limited. A complex interplay between heat cytotoxicity, systemic inflammatory response, and disseminated intravascular coagulation causes multiple organ dysfunctions. Additionally, brain hyperthermia causes CNS abnormalities, increasing metabolic rate with a reduction in blood flow in the cerebrum [21]. Although some anticoagulant agents such as recombinant activated protein C [22,23], recombinant thrombomodulin [18], and antithrombin [19] attenuate inflammation and multi-organ system dysfunction in animal experiments, the clinical syndrome of heatstroke is still a challenging problem, and there remains an urgent unmet need for development of new therapeutics. We demonstrated for the first time, to our knowledge, that cholinergic activation by electrical VNS attenuated heat stress-related systemic inflammation, resulting in an increased survival rate in a rat model of heatstroke.
Systemic inflammation plays a pivotal role in the pathogenesis of heatstroke. A variety of cytokines are known to be produced in response to endogenous or environmental heat stress [24]. It is now well accepted that signaling from the vagus nerve has potent anti-inflammatory effects [25]. Electrical VNS against a burn injury model reduced serum levels of TNF-a and IL-6 [26,27], and in a traumatic brain injury model, the suppressive effect of VNS on TNF-a was shown in intestine [28]. In the present study, we also showed that induction of serum TNF-a and IL-6 was significantly suppressed by electrical VNS treatment in our rat model of heatstroke. In addition, heat stress enhanced the number of inflammatory cells in the lung and the spleen, and VNS treatment hampered it. The obvious and rapid decline in the levels of these mediators by VNS may have contributed to the reduction in multiple organ dysfunctions, resulting in improvement of outcome in the heatstroke rats.
The endothelium is an interactive barrier between blood and tissue that plays a critical role in host immune responses in systemic inflammatory response syndrome. Extreme heat stress directly induces endothelial death by apoptosis [29]. There have been several reports concerning the effect of cholinergic activation against endothelial injury. Acetylcholine and other cholinergic mediators inhibit the expression of E-selectin in endothelial cells stimulated by IL-4 [30]. Another study provided evidence that pharmacologic stimulation of a7nAChR reduces both chemokine production and adhesion molecule expression in endothelium during inflammation [31]. The present study showed that electrical VNS significantly suppressed serum levels of endothelial injury markers sTM and sE-selectin, suggesting that VNS protects endothelium from heat stress. Because widespread hemorrhage, thrombosis, and transmural migration of leukocytes in association with microvascular endothelial injury are prominent features of heatstroke [29], it is challenging to elucidate the regulatory mechanism of the vagus nerve on endothelial cellular function.
a7nAChRs, an important target of the cholinergic antiinflammatory pathway, are known to be expressed on immune cells, especially macrophages, and endothelial cells [32]. The production of pro-inflammatory cytokines derived from these cells is restricted through stimulation of a7nAChRs [33]. Huston et al. [34] reported that electrical VNS failed to inhibit systemic TNF production in splenectomized mice during lethal endotoxemia, indicating that the spleen is a major contributor to the antiinflammatory effect via the cholinergic pathway. Exposure of inflammatory cells having a7nAChR to acetylcholine in spleen is an essential step to suppress systemic inflammation [6]. To clarify the role of the cholinergic anti-inflammatory pathway in the pathogenesis of heatstroke, we evaluated the expression of a7nAChR on the inflammatory cells. Most of the infiltrated CD11b-postive cells expressed a7nAChR in the lung and not in the spleen. The spleen stores undifferentiated monocytes readily recruitable to augment inflammation at distant sites. After the release of splenic monocytes, they become biologically active when recruited to inflamed organs [35]. It is possible that the enhanced expression of a7nAChR on infiltrated inflammatory cells and endothelial cells in the lung indicates a compensatory reaction to maintain immune homeostasis. However, heat stress would disturb the neural output signal to stimulate a7nAChR and consequently abolish this intrinsic immunomodulatory mechanism. In addition, the infiltrated inflammatory cells in spleen would be held in an immature state and might be unable to produce substantial a7nAChR in the acute phase of heatstroke. Further investigation is necessary to support our supposition.
Lastly, in the present study, electrical VNS treatment was conducted immediately after the heat stress. The time schedule of treatment in this experimental study may differ from the clinical condition, and this is a limitation of the study. Therefore, further investigation including the effects of delayed VNS treatment on heatstroke is required. In conclusion, we demonstrated in a rat model of heatstroke that electrical VNS suppressed systemic inflammation occurring secondary to heat stress and eventually improved survival. The present study defined the role of the cholinergic anti-inflammatory pathway in the pathophysiology of heatstroke. These findings may provide a novel therapeutic strategy using electrical VNS to combat severe heatstroke in the critical care setting.