INTRODUCTION

Every year, about 50 million people worldwide suffer from sepsis, and about 11 million people die from it (1). In 2016, the Third International Consensus Definition of Sepsis and Septic Shock (Sepsis-3) defined septic shock as a subset of sepsis that increases mortality due to abnormalities in circulation, cellular function, and metabolism (2). Norepinephrine (NE) is recommended in Sepsis-3 as a vasopressor agent for septic shock, which is a circulatory condition with reduced peripheral vascular resistance (2). However, we often experience peripheral tissue circulatory failure in real-world clinical practice with the use of NE. Nonocclusive mesenteric ischemia (NOMI) exists as one element of peripheral circulatory failure, and we have seen cases of NOMI associated with NE administration for septic shock (3). In a previous study in our department, we showed that administration of NE could induce ischemia in mesenteric tissue in a rabbit model of endotoxic shock (4). These results suggest the need for drugs with a pressure-raising effect that can also maintain peripheral circulation, including intestinal mucosal blood flow, in septic shock.

Therefore, we searched for an alternative drug for endotoxic shock and focused on 2-methyl-2thiazoline (2MT), which has an inhibitory effect on inflammatory cytokines. 2-Methyl-2thiazoline is one of the thiazoline-related fear odors, which were developed as a group of artificial odor molecules that induce innate fear-related behaviors in animals (5). It also induces integrated physiological responses, including anti-inflammatory, hypometabolic, and hypoxic resistance responses, which may increase the probability of survival in crisis conditions and has been shown to exert therapeutic effects in animal models of ischemia-reperfusion and hypoxic injury (6–8). In mouse models, 2MT stimulation together with LPS has been shown to have a strong anti-inflammatory effect and to improve survival rate (9) and is expected to be clinically applied to the treatment of sepsis.

No previous studies have examined the effect of 2MT dosing on circulatory dynamics, which remains unproven. Therefore, the hypothesis of this study is that the hemodynamics of LPS-induced endotoxic shock would be ameliorated by the administration of 2MT. The purpose of this study was thus to evaluate the temporal changes in superior mesenteric venous (SMV) blood flow and jejunal mucosal tissue blood flow in the intestinal vasculature, and in systematic hemodynamics, caused by the administration of 2MT in both normal rabbits and those with LPS-induced shock.

MATERIALS AND METHODS Surgical preparation

This study was approved by the Animal Care Committee of Kansai Medical University (approval number, 21-121) and conformed to the US National Institutes of Health standards for animal experimentation. Twenty-four New Zealand White rabbits weighing 2.70 to 3.53 kg each were inductively anesthetized with midazolam (0.33 mg/kg i.v.), medetomidine hydrochloride (0.08 mg/kg i.v.), and butorphanol tartrate (0.08 mg/kg i.v.), followed by constant infusion of pentobarbital sodium at 10 mg/kg/h, ketamine hydrochloride (10 mg/kg i.m., followed by 5 mg/kg i.m. every 60 min). As a maintenance solution and anesthetic carrier, normal saline was intravenously infused at a constant rate of 10 mL/kg/h throughout the experiment. For measurement of arterial pressure, a 22-gauge low-compliance catheter was inserted into the left ear artery. Tracheostomy was performed; a 4-mm endotracheal tube, which we prepared ourselves, was inserted; and the tube was fixed with a 2-0 silk ligature. A thermistor probe (Transonic Flowprobe; Transonic Systems Inc, Ithaca, NY) for measuring cardiac output (CO) was inserted into the carotid artery. A 3-Fr Atom (43550; Atom Medical, Tokyo, Japan) indwelling feeding tube was inserted into the jugular vein to infuse 1 mL of cold saline, for measurement of CO by the thermodilution method. Cardiac output values were converted to cardiac index (CI) values by dividing CO by body weight (mL/min/kg). After laparotomy was performed via a transverse epigastric incision, the superior mesenteric vein was carefully dissected from the perivascular tissue, and a flow probe was placed around it. A mesenteric arcade dominated by a single artery was selected, and a loop with a length of about 50 mm in the jejunum was defined. Both ends of the jejunal loop were ligated with 2-0 silk to block any collateral blood circulation. Then, the jejunal loop was opened with electrocoagulation on the contralateral side of the mesentery and exteriorized on the ventral side of the rabbit. The jejunum was placed mucosal surface up on a handmade platform covered with black cellophane, and while only handling the serous membrane of the incision with forceps, we fixed it with cyanoacrylate glue. A transparent low-compliance plastic adhesive film was then placed on the cellophane to cover the entire surface of the jejunal mucosa and was externally fitted with minimal compression. Care was taken not to leave any air bubbles between the mucosal surface and the plastic film. The abdominal wound was closed with 3-0 nylon, leaving a window large enough to prevent strangulation of the jejunal mesentery. After surgery, the rabbits were started on rocuronium bromide (7 μg/kg/min i.v.). The rabbits were ventilated mechanically with a Harvard nonrebreathing ventilator (Harvard Apparatus, Holliston, MA) at a frequency adjusted to set the baseline arterial PCO2 to a target level of 35 to 40 mm Hg, with a tidal volume of 25 mL (FiO2, 50%). The ventilatory frequencies were not changed thereafter. Rabbits were maintained at a constant body temperature of 39.0°C to 40.0°C on a heated operating table (Model 59-6,833; Harvard Apparatus).

Determination of mucosal blood flow

A laser Doppler imager scanner (LDIS) (Moor LDI2; Moor Instruments Ltd, Axminster, Devon, United Kingdom) was used to measure mucosal blood flow in the jejunum. In brief, the LDIS uses a computer-controlled mirror to scan the tissue surface with a low-power laser beam. The laser beam passes through the tissue, and some of the incident light is scattered by the moving red blood cells, broadening the frequency range. The imager detects some of this scattered light and processes it electronically and digitally to display a color-coded image of blood flow. The LDIS records flux measurements of mucosal blood flow, usually measured in perfusion units (PU), for every few thousands of pixels that correspond to the scanned area. Mucosal blood flow was measured in a rectangular area of approximately 300 pixels of interest longitudinally located on the mesenteric side of the scanned surface image using Moor LDI software. The average flux and heterogeneity of the blood flow were evaluated by the mean and coefficient values of measured PU at each pixel in the region of interest, respectively.

Experimental protocol

After surgery, the rabbits were stabilized for 60 min, and a baseline data set was obtained. During this stabilization period, additional intravenous saline was administered at 20 mL/h for 1 h to replace ascites and exudate lost during surgery. For the baseline values (time 0), we measured MAP, heart rate (HR), CO (Model ML845 ML846; AD Instruments, Castle Hill, Australia), SMV blood flow (Model TS420; Transonic Systems Inc, New York, United States), and jejunal mucosal tissue blood flow (Moor LDI2; Moor Instruments Ltd) and took an arterial blood sample to measure the lactate acid (Lac) level. We measured CO three times by the thermodilution method (with 1 mL of saline solution injected at 0°C), and the average value was calculated. The animals were then randomly placed into the (1) control group (n = 6), (2) LPS (Escherichia coli O111, 1 mg/kg) group (n = 6), (3) 2MT (80 mg/kg) group (n = 6), or (4) LPS-2MT group (n = 6). After measuring baseline values, we intravenously administered LPS (1 mg/kg) to the LPS group and LPS-2MT group for 5 min. After baseline measurement in the 2MT group and after LPS administration in the LPS-2MT group, 2MT (80 mg/kg) was administered intravenously for 5 min. After administration of the LPS and 2MT, we measured MAP, CO, SMV blood flow, and jejunal mucosal tissue blood flow and took an arterial blood sample to measure lactate level every 30 min from 0 to 240 min. At the end of the experiment, the rabbits were euthanized by overdose of anesthetic.

Statistical analysis

All values are expressed as mean ± SD. Superior mesenteric venous blood flow and jejunal mucosal blood flow were analyzed as percent change with the baseline set at 1. First, we performed a t test on the control and 2MT groups to confirm the effect of 2MT monotherapy. Second, we performed two-way ANOVA of LPS (+/−) and 2MT (+/−) for all measurement items. A P value of <0.05 was considered statistically significant. All statistical analyses were performed using IBM SPSS version 26 (SPSS, Inc, Chicago, IL).

RESULTS

First, to understand the effect of 2MT monotherapy, we present the results for each measured item for the control and 2MT groups (Table 1). We performed t tests on each item and found significant differences between the values of MAP at 30 to 150 min and for Lac at 60 to 120 min. Next, we present the results for each measured item across the four groups. A two-way ANOVA of LPS (+/−) and 2MT (+/−) was performed to show the interaction between LPS and 2MT.

Table 1:

Actual values of MAP, HR, CI, and Lac level from baseline to 240 min

Mean arterial pressure

In the control group, MAP was maintained until 240 min. In the LPS group, it decreased by 45% and was maintained until 240 min. In the 2MT group, it increased by 25% immediately after 2MT administration, gradually decreased by 150 min to a level similar to that of the control group, and was maintained until 240 min. In the LPS-2MT group, it increased by 25% immediately after LPS and 2MT administration, took 60 min to decrease to the level of the control group, and then gradually decreased but remained at about 80% of the value at baseline until 240 min. Two-way ANOVA showed a significant interaction between LPS and 2MT at 210 min for MAP (Fig. 1).

Fig. 1:

Mean arterial pressure from baseline to 240 min. Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed a significant interaction between LPS and 2MT for MAP at 210 min. 2MT, 2-methyl-2-thiazoline.

Heart rate

In the control group, HR increased by 10% after 150 min. In the LPS group, no significant change in HR was observed. In the 2MT group, HR increased by 15% of baseline at 60 to 90 min and was maintained up to 240 min. In the LPS-2MT group, HR increased by about 15% of baseline at 90 to 120 min and was maintained up to 240 min. Two-way ANOVA showed no significant interaction between LPS and 2MT for HR (Fig. 2).

Fig. 2:

Heart rate from baseline to 240 min. Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed no significant interaction between LPS and 2MT for HR. 2MT, 2-methyl-2-thiazoline; HR, heart rate.

Confidence interval

In the control and 2MT groups, CI decreased slightly over time but was maintained until 240 min. In the LPS group, CI decreased by 60% until 60 min and then maintained a similar value thereafter. In the LPS-2MT group, CI dropped to about 65% of baseline until 60 min but then recovered after 90 min to a level similar to that of the control group and was maintained until 240 min. Two-way ANOVA showed a significant interaction between LPS and 2MT for CI at 120, 180, and 210 min (Fig. 3).

Fig. 3:

Cardiac index from baseline to 240 min. Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed a significant interaction between LPS and 2MT for CI at 120, 180, and 210 min. 2MT, 2-methyl-2-thiazoline; CI, cardiac index.

SMV blood flow

In the control and 2MT groups, SMV blood flow was maintained until 240 min. In the LPS group, SMV blood flow transiently decreased to 60% of the baseline level early after LPS administration but then rose to a level about 110% of baseline after 120 min. In the LPS-2MT group, SMV blood flow showed a transient decrease to 60% of baseline early after LPS administration but then increased to about 110% of baseline after 90 min. Two-way ANOVA showed a significant interaction between LPS and 2MT for SMV blood flow at 90 min (Fig. 4).

Fig. 4:

Superior mesenteric venous blood flow from baseline to 240 min. The values shown are based on the ratio of change from the baseline level (mean of the experimental actual value/baseline level). Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed a significant interaction between LPS and 2MT for SMV blood flow at 90 min. 2MT, 2-methyl-2-thiazoline; SMV, superior mesenteric venous.

Jejunal mucosal tissue blood flow

In the control and 2MT groups, jejunal mucosal tissue blood flow decreased gradually until 240 min but remained at about 80% of baseline. In the LPS group, it decreased temporarily to 40% of baseline between 60 and 90 min and then improved to 60% of the baseline value. In the LPS-2MT group, jejunal mucosal tissue blood flow decreased to 60% of baseline in the first 60 min but gradually improved to 90% of the control group value by 120 min and remained there. Two-way ANOVA showed a significant interaction between LPS and 2MT for jejunal mucosal tissue blood flow at 90 to 180 and 240 min (Fig. 5).

Fig. 5:

Jejunal mucosal tissue blood flow from baseline to 240 min. The values shown are based on the ratio of change from the baseline level (mean of the experimental actual value/baseline level). Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed significant interaction between LPS and 2MT for jejunal mucosal tissue blood flow at 90 to 180 and 240 min. 2MT, 2-methyl-2-thiazoline.

Lactate

In the control and 2MT groups, Lac was maintained at a slight increase until 240 min. In the LPS group, a prominent increase in Lac was seen at the same time as the decrease in blood pressure, and Lac was maintained at a high level. In the LPS-2MT group, the increase in Lac was similar to that in the LPS group until 90 min, but further increases were suppressed thereafter. Two-way ANOVA showed a significant interaction between LPS and 2MT for IVC blood flow at 180 to 240 min (Fig. 6).

Fig. 6:

Lactic acid level from baseline to 240 min. Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed a significant interaction between LPS and 2MT for Lac after 180 min. 2MT, 2-methyl-2-thiazoline; Lac, lactic acid.

SMV blood flow/CO ratio

In the control and 2MT groups, the SMV blood flow/CO ratio was maintained until 240 min. In the LPS group, it began to increase after 90 min and remained at 180% of the reference value. In the LPS-2MT group, it increased after 90 min and remained at 130% of the reference value. Two-way ANOVA showed a significant interaction between LPS and 2MT for the SMV blood flow/CO ratio at 120, 210, and 240 min (Fig. 7).

Fig. 7:

Ratio of superior mesenteric venous blood flow to CO from baseline to 240 min. Control group (n = 6), LPS group (n = 6), 2MT group (n = 6), and LPS-2MT group (n = 6). Two-way ANOVA showed a significant interaction between LPS and 2MT in SMV blood flow/CO ratio at 120, 210, and 240 min. 2MT, 2-methyl-2-thiazoline; CO, cardiac output; SMV, superior mesenteric venous.

DISCUSSION

The present results showed that administration of 2MT in an endotoxic shock model maintained MAP and restored SMV blood flow while also maintaining jejunal mucosal tissue blood flow. In the rabbit endotoxic shock model, systematic hemodynamics showed a prolonged decrease in MAP mainly because of the reduction in CI with an increase in Lac after LPS administration, indicating a state of hypodynamic shock. In the 2MT group compared with the control group, although the difference was not significant, the increase in MAP without an increase in CI indicated that 2MT has a vasoconstrictor-like effect. This was confirmed by the result that, in the LPS-2MT group at 30 min, CI was decreased but MAP was increased. The combination of LPS and 2MT predominantly restored hypodynamic shock after 90 min, indicating that 2MT significantly improved, albeit only partially, the shock induced by LPS.

We then focused on the intestinal circulation system in terms of abnormal vascular resistance. In the LPS group, although SMV blood flow showed a transient decrease, it then improved from baseline values, but the CI continued to decrease, causing an increase in the SMV blood flow/CO ratio. This suggests that abnormal vascular resistance in the splanchnic circulation is one factor involved in the abnormal hemodynamics seen in endotoxic shock. At the same time, jejunal mucosal tissue blood flow reached a minimum at 90 min after LPS administration and then gradually recovered to about 60% of baseline. This also suggests that blood flow in the splanchnic circulation increases but that ischemic change in the mucosa occurs in endotoxic shock. Although no effect on jejunal mucosal blood flow was observed in the 2MT group, administration of 2MT to the rabbit endotoxic shock model improved SMV and jejunal mucosal blood flow with recovery of both the CI and the SMV blood flow/CO ratio. Thus, 2MT may improve abnormal vascular resistance in the splanchnic circulation induced by LPS administration. These findings indicate that, in the rabbit endotoxic shock model with LPS administration, 2MT improves the CI and MAP of the systemic and intestinal circulation but does not cause excessive reduction of peripheral vascular resistance and may also improve jejunal mucosal tissue blood flow.

Our previous study showed that the administration of NE in an LPS-induced endotoxic shock model maintained SMV blood flow but markedly reduced intestinal mucosal blood flow because of the interaction of LPS and NE (4). It is expected that 2MT can reduce the occurrence of complications of peripheral circulatory failure that may occur with NE administration and improve circulatory dynamics.

Tracey (9) reported that electrical stimulation of the vagus nerve suppressed inflammatory cytokines and reduced lethality in a sepsis model. Matsuo et al. (10) showed that 2MT binds to transient transient receptor potential ankyrin 1 (TRPA1) in the trigeminal and vagus nerves, activating the brainstem to midbrain pathway and inducing hypometabolism and hypoxic resistance. They also showed that, in a mouse model of endotoxic shock, 2MT administration suppressed the secretion of proinflammatory cytokines and enhanced the secretion of anti-inflammatory cytokines, such as IL-10, leading to a decrease in mortality (11). It is known that inflammatory cytokines induced with the administration of LPS decrease CO and peripheral vascular resistance. Therefore, we hypothesized that the improved survival of the endotoxic shock model with 2MT administration would likely be due in part to improved circulatory dynamics. In the present study, we showed that 2MT administration not only maintained MAP and restored SMV blood flow while maintaining jejunal mucosal tissue blood flow but also improved lactate levels, which are considered an indicator of anaerobic metabolism during peripheral circulatory failure. These results strongly support our hypothesis that 2MT improves circulatory dynamics. Suppression of inflammatory cytokines may improve the state of increased vascular permeability, leading to maintenance of circulating blood volume and improvement of the CI observed in this study.

As well, we also observed an increase in blood pressure immediately after 2MT administration. The detailed mechanism behind the effect of 2MT to elevate pressure in the rabbit endotoxic shock model and on rabbits not receiving LPS is not known. We assume that 2MT may have some effects on the autonomic nervous and cardiovascular systems. Subcutaneous administration of 2MT has been shown to have no effect on the rate of left ventricular inner diameter shortening or left ventricular ejection fraction (8). Other than this, none of the literature we could find reported a direct assessment of the effect of 2MT on cardiac function. Further research is needed on the impact of 2MT on the circulation.

In addition, intestinal mucosal blood flow, which is decreased during endotoxic shock, was shown to improve with the administration of 2MT in this study. In recent years, there has been a growing interest in elucidating the role of TRP channels in gastrointestinal physiology, including intestinal motility, secretion, and visceral sensation (12–15). Kono et al. (16) reported that, in their assessment of the effect of TRPA1 on intestinal blood flow, the TRPA1 agonist allyl-isothiocyanate (0.002 mg/kg) slowly increased vasodilation, which peaked after 120 min. Therefore, it is possible that 2MT improved jejunal mucosal tissue blood flow not only through an indirect pathway via sensory information processing by the brain center but also through a direct pathway via TRPA1 in the gut.

To our knowledge, this study is the first description of experiments in which 2MT was administered to a rabbit endotoxic shock model, which has similar circulatory dynamics to those of septic shock, and circulatory dynamics could be observed. Norepinephrine, the first-line drug for circulatory management in septic shock, has a hypertensive effect but does not improve intestinal mucosal blood flow. However, the administration of 2MT to a rabbit endotoxic shock model could maintain intestinal circulation in addition to providing its hypertensive effect. Although it should be fully considered that the pathogenesis of endotoxic shock in humans is different from that of endotoxic shock in this experiment, this finding raises the possibility of treating septic shock without fatal complications such as NOMI, which is one element of peripheral circulatory failure.

Limitations

First, we used different anesthetics in this experiment than in our previous experiments, and the circulatory dynamics reproduced by the rabbit endotoxic shock model were in a hypodynamic state. However, the model reproduced the same general principles as in the previous experiment. Second, inflammatory mediators that could be related to the vascular responses observed in this study were not measured. Third, our findings were obtained with an LPS model as the model of sepsis.

CONCLUSION

We showed the efficacy of 2MT against a rabbit model of endotoxic shock. The results of this study will help clinicians in the future if the potential of 2MT as a new treatment for septic shock can be realized.

ACKNOWLEDGMENTS

The authors thank Mr Tomoki Kitawaki (Mathematics Laboratory, Kansai Medical University) for helpful suggestions and insights on statistical processing.

REFERENCES 1. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, Colombara DV, Ikuta KS, Kissoon N, Finfer S, et al.: Global, regional, and national sepsis incidence and mortality, 1990–2017: analysis for the Global Burden of Disease Study. Lancet 395(10219):200–211, 2020. 2. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche JD, Coopersmith CM, et al.: The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 315:801–810, 2016. 3. Nakamura F, Yui R, Muratsu A, Onoe A, Nakajima M, Takahashi H, Kishimoto M, Sakuramoto K, Muroya T, Kajino K, et al.: A strategy for improving the prognosis of non-occlusive mesenteric ischemia (NOMI): a single-center observational study. Acute Med Surg 6(4):365–370, 2019. 4. Nakamura F, Muroya T, Onoe A, Ikegawa H, Kuwagata Y: Effects of norepinephrine on the intestinal vascular system in rabbits with endotoxic shock. Shock 55(6):827–831, 2021. 5. Isosaka T, Matsuo T, Yamaguchi T, Funabiki K, Nakanishi S, Kobayakawa R, Kobayakawa K: Htr2a-expressing cells in the central amygdala control the hierarchy between innate and learned fear. Cell 163(5):1153–1164, 2015. 6. Wang Y, Cao L, Lee CY, Matsuo T, Wu K, Asher G, Tang L, Saitoh T, Russell J, Klewe-Nebenius D, et al.: Large-scale forward genetics screening identifies Trpa1 as a chemosensor for predator odor-evoked innate fear behaviors. Nat Commun 9(1):2041, 2018. 7. Matsuo T, Isosaka T, Tang L, Soga T, Kobayakawa R, Kobayakawa K: Artificial hibernation/life-protective state induced by thiazoline-related innate fear odors. Commun Biol 4(1):101, 2021. 8. Nishi M, Ogata T, Kobayakawa K, Kobayakawa R, Matsuo T, Cannistraci CV, Tomita S, Taminishi S, Suga T, Kitani T, et al.: Energy-sparing by 2-methyl-2-thiazoline protects heart from ischaemia/reperfusion injury. ESC Heart Fail 9(1):428–441, 2022. 9. Tracey KJ: The inflammatory reflex. Nature 420:853–859, 2002. 10. Matsuo T, Isosaka T, Hayashi Y, Tang L, Doi A, Yasuda A, Hayashi M, Lee CY, Cao L, Kutsuna N, et al.: Thiazoline-related innate fear stimuli orchestrate hypothermia and anti-hypoxia via sensory TRPA1 activation. Nat Commun 12(1):2074, 2021. 11. Matsuo T, Isosaka T, Tang L, Soga T, Kobayakawa R, Kobayakawa K: Thiazoline-related TRPA1 agonist odorants orchestrate survival fate in mice. bioRxiv . doi:org/10.1101/2020.05.17.100933. 12. Kaji I, Karaki S, Kuwahara A: Effects of luminal thymol on epithelial transport in human and rat colon. Am J Physiol Gastrointest Liver Physiol 300(6):G1132–G1143, 2011. 13. Kaji I, Yasuoka Y, Karaki SI, Kuwahara A: Activation of TRPA1 by luminal stimuli induces EP4-mediated anion secretion in human and rat colon. Am J Physiol Gastrointest Liver Physiol 302(7):G690–G701, 2012. 14. Nozawa K, Kawabata-Shoda E, Doihara H, Kojima R, Okada H, Mochizuki S, Sano Y, Inamura K, Matsushime H, Koizumi T, et al.: TRPA1 regulates gastrointestinal motility through serotonin release from enterochromaffin cells. Proc Natl Acad Sci U S A 106:3408–3413, 2009. 15. Venkatachalam K, Montell C: TRP channels. Annu Rev Biochem 76:387–417, 2007. 16. Kono T, Kaneko A, Omiya Y, Ohbuchi K, Ohno N, Yamamoto M: Epithelial transient receptor potential ankyrin 1 (TRPA1)-dependent adrenomedullin upregulates blood flow in rat small intestine. Am J Physiol Gastrointest Liver Physiol 304(4):G428–G436, 2013.