GTS-21 reduces microvascular permeability during experimental endotoxemia

Accepted Manuscript GTS-21 reduces microvascular permeability during experimental endotoxemia

Karsten Schmidt, Sukanya Bhakdisongkhram, Florian Uhle, Christoph Philipsenburg, Aleksandar R. Zivkovic, Thorsten Brenner, Johann Motsch, Markus A. Weigand, Stefan Hofer PII: DOI: Reference:

S0026-2862(17)30063-8 doi: 10.1016/j.mvr.2017.08.002 YMVRE 3729

To appear in:

Microvascular Research

Received date: Revised date: Accepted date:

24 March 2017 3 July 2017 12 August 2017

Please cite this article as: Karsten Schmidt, Sukanya Bhakdisongkhram, Florian Uhle, Christoph Philipsenburg, Aleksandar R. Zivkovic, Thorsten Brenner, Johann Motsch, Markus A. Weigand, Stefan Hofer , GTS-21 reduces microvascular permeability during experimental endotoxemia, Microvascular Research (2017), doi: 10.1016/ j.mvr.2017.08.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Title GTS-21 reduces microvascular permeability during experimental endotoxemia Karsten Schmidt, MD*, Sukanya Bhakdisongkhram, MD*, Florian Uhle, PhD*, Christoph Philipsenburg, MD*, Aleksandar R. Zivkovic, MD*, Thorsten Brenner, MD*,

PT

Johann Motsch, MD*, Markus A. Weigand*, MD, Stefan Hofer, MD†

RI

*Department of Anesthesiology, Heidelberg University Hospital

SC

Im Neuenheimer Feld 110, 69120 Heidelberg, Germany †

NU

Clinic for Anesthesiology, Intensive Care and Emergency Medicine I, Westpfalz

MA

Hospital, Kaiserslautern, Germany

D

Corresponding author:

PT E

Karsten Schmidt, MD

E-mail: [email protected]

CE

Phone: +49 (0) 6221-56 37840

AC

Fax: +49 (0) 6221-56 53 45 E-mail address of the co-authors: Sukanya Bhakdisongkhram

[email protected]

heidelberg.de Florian Uhle

[email protected]

Aleksandar R. Zivkovic

[email protected]

Christoph Philippsenburg

[email protected] 1

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

[email protected]

Johann Motsch

[email protected]

Markus A. Weigand:

[email protected]

Stefan Hofer:

[email protected]

PT

Thorsten Brenner:

Authors’ contributions: KS designed the study, supervised the data acquisition and

RI

wrote the manuscript. KS, SB, FU performed the statistical analysis. SB carried out

SC

the experiments. CP supervised the experiments. SB, FU, CP, ARZ, TB, JM, MAW

NU

and SH have been involved in drafting the manuscript and revising it critically. All authors read and approved the final manuscript.

MA

Funding

AC

CE

PT E

D

This project was funded by the B. Braun-Stiftung, Melsungen, Germany.

2

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Abstract Introduction: No effective pharmacological therapy is currently available to attenuate tissue edema formation due to increased microvascular permeability in sepsis. Cholinergic

PT

mediators have been demonstrated to exert anti-inflammatory effects via the α7 nicotinic acetylcholine receptor (α7nAChR) during inflammation. GTS-21, a partial

RI

α7nAChR agonist, is an appealing therapeutic substance for sepsis-induced

SC

microvascular inflammation due to its demonstrated cholinergic anti-inflammatory properties and its favorable safety profile in clinical trials. This study evaluated the

NU

effect of GTS-21 on microvascular permeability and leukocyte adhesion during

MA

experimental endotoxemia. Methods:

D

Male Wistar rats (n=60) were anesthetized and prepared for intravital microscopy

PT E

(IVM). Sevoflurane inhalation combined with propofol (10mg/kg) and fentanyl (5μg/kg) was used for anesthesia induction, followed by continuous intravenous anesthesia with propofol (10-40 mg/kg/h) and fentanyl (10μg/kg/h). The rat

CE

mesentery was prepared for evaluation of macromolecular leakage, leukocyte

AC

adhesion and venular wall shear rate in postcapillary venules using IVM. Following baseline IVM recording, GTS-21 (1mg/kg) was applied simultaneously with, 1h prior to and 1h after administration of lipopolysaccharide (LPS, 5 mg/kg). Test substances (crystalloid solution, LPS, GTS-21) were administered as volume equivalent intravenous infusions over 5 min in the respective treatment groups. The consecutive IVMs were performed at 60, 120 and 180 min after the baseline IVM. The systemic inflammatory response was evaluated by measuring TNF-ɑ levels after the 180 min IVM. 3

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Results: Microvascular permeability was significantly reduced in animals treated with GTS-21 simultaneously and 1h after induction of endotoxemia. Leukocyte adhesion, venular wall shear rate and TNF-ɑ levels were not affected by GTS-21 treatment compared to

PT

the untreated endotoxemic animals. Conclusion:

RI

GTS-21 has a protective effect on microvascular barrier function during endotoxemia.

SC

Considering its anti-inflammatory efficacy and safety profile, its clinical use might

AC

CE

PT E

D

MA

NU

prove beneficial for the treatment of capillary leakage in sepsis therapy.

4

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Introduction Sepsis-induced microvascular inflammation is characterized by increased endothelial permeability with concomitant tissue edema formation and leukocyte adhesion. Consequently, sepsis leads to a functional breakdown of the microcirculation with

PT

disrupted microvascular barrier function and impaired tissue oxygenation. The severity of these microcirculatory alterations is closely associated with sepsis

RI

mortality (De Backer et al., 2014). The activation of cholinergic neuro-immunological

SC

mechanisms has been demonstrated to modulate the magnitude of the innate

NU

immune response by limiting pro-inflammatory processes to a non-harmful degree, thereby minimizing tissue injury (Tracey, 2007; Tracey, 2009). Acetylcholine (ACh) is

MA

the primary transmitter for cholinergic anti-inflammatory signaling and activates a variety of different cholinergic receptors. In the context of immune cells, the homo-

D

pentameric nicotinic receptor consisting of subunit α7 (α7nAChR) has been shown to

PT E

mediate these anti-inflammatory effects (de Jonge and Ulloa, 2007; Pohanka, 2012). Subsequent intracellular signal transduction inhibits the transcription of pro-

CE

inflammatory genes (Tracey, 2007; Tracey, 2009). GTS-21 is a partial α7nAChR agonist that has been evaluated in clinical trials for neurodegenerative diseases

AC

(Conejero-Goldberg et al., 2008; Kem, 2000). The anti-inflammatory effect of GTS-21 has been documented in rodent and human inflammation models in vitro and in vivo (Cai et al., 2009; Chatterjee et al., 2009; Conejero-Goldberg et al., 2008; Giebelen et al., 2007; Khan et al., 2012; Kox et al., 2011a; Kox et al., 2011b; Kox et al., 2009; Nullens et al., 2016; Pavlov et al., 2007). Human and rat endothelial cells express α7nAChR and the endothelium has been identified as a target for anti-inflammatory cholinergic mediators (Chatterjee et al., 2009; Kirkpatrick et al., 2001; Moccia et al., 2004; Saeed et al., 2005). Chatterjee et al. showed that inflammatory endothelial 5

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

activation can be ameliorated in vitro by GTS-21 treatment in human macrovascular and microvascular endothelial cells (Chatterjee et al., 2009). However, no effective pharmacological therapy is currently available to restore microvascular endothelial barrier function and ameliorate capillary leakage in human sepsis. Our group

endothelial

barrier

function

in

previously

PT

demonstrated beneficial cholinergic anti-inflammatory effects on microvascular published

intravital

microscopic

RI

investigations (Peter et al., 2010; Schmidt et al., 2015a). Therefore, we investigated

SC

the in-vivo effect of the partial α7nAChR-agonist GTS-21 on microvascular inflammation during LPS-induced endotoxemia in rats using an experimental IVM

MA

NU

model.

Materials and Methods

D

Animals

PT E

All experimental procedures were performed under the German legislation on animal protection and approved by the Governmental Animal Protection Committee (project

CE

licence number 35-9185.81/G-185/15). Male Wistar rats (wild type, 7-11 weeks old, body weight: 342.3 ± 39.5 g, n=60, strain: RjHan:WI; Janvier; St Berthevin, France)

AC

were used in this study. Groups of three rats were housed in stainless steel cages under standardized conditions (12-h light-dark cycle, 21± 2°C, 40%–50% humidity) with unlimited access to water and regular commercial chow. The animals were allowed to acclimatize in a specific pathogen-free animal facility at the Interfacultary Biomedical Faculty of Heidelberg University one week before the experiments were performed

in

the

intravital

microscopy

laboratory

of

the

Department

of

Anesthesiology, Heidelberg University Hospital.

6

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Anesthesia and animal preparation During the experiments, sufficient anesthetic depth and maintenance was thoroughly monitored (regular tests of the pedal withdrawal reflex of the hind limb and the corneal reflex, observance of ventilator tolerance, continuous heart rate and blood

PT

pressure monitoring). All investigators involved had been specially trained for rodent anesthesia. Atraumatic surgical preparation of all animals was performed by one

RI

investigator within 15 min. To reduce the stress response to surgery, sufficient

SC

anesthetic depth was confirmed before each consecutive operative step.

NU

First, animals were anesthetized using sevoflurane (Sevorane®, AbbVie Deutschland GmbH & Co.KG, Ludwigshafen, Germany) inhalation. The left femoral artery was

MA

cannulated for arterial blood gas analysis collection and for continuous real-time monitoring of mean arterial pressure (MAP [mmHg]) and heart frequency (HF [min -

D

1]). The left femoral vein was cannulated next and 5μg/kg fentanyl (Fentanyl®,

PT E

Janssen-Cilag GmbH, Neuss, Germany) and 10mg/kg propofol (Propofol 2%, 20 mg/ml, Fresenius Kabi Deutschland GmbH, Bad Homburg, Germany) were

CE

administered.

AC

Following atraumatic intubation with a G 14 puncture needle catheter from a Cavafix® Certo® Code 335 set (B. Braun Melsungen AG, Hospital Care, 34209 Melsungen, Deutschland), cannulation of the right jugular vein and midline laparotomy were performed. After the intubation sevoflurane was discontinued, the animals were mechanically ventilated with a rodent ventilator and a total intravenous anesthesia regime was used for the rest of the experiment. The left femoral vein catheter was used for test substance and maintenance fluid administration and for blood sample collection. The right jugular vein catheter was 7

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

used for continuous intravenous anesthesia with fentanyl 10μg/kg/h and propofol 1040 mg/kg/h. The average propofol maintenance dose used in this study was 17 mg/kg/h (minimum dose 11.7 mg/kg/h, maximum dose 35.1 mg/kg/h). If required, analgesia and anesthesia were deepened with a fentanyl bolus (5μg/kg) followed by

PT

a propofol bolus (5mg/kg) if the fentanyl effect remained insufficient. All catheters were flushed once with heparin (50 IU) before insertion to avoid catheter occlusion.

RI

After the surgical preparation, the animals were placed on a microscopy stage and a

SC

segment of the ileum was gently exteriorized through the mid-line abdominal incision.

NU

The mesentery was carefully draped over a clear glass pedestal and areas of interest with macroscopically visible vascular beds were identified. In the following 45 min

MA

stabilization period no further manipulations or microscopic measurements were performed to reduce leukocyte activation and endothelial leakage associated with the

D

surgical preparation.

PT E

Vital sign monitoring, maintenance fluid regime, ventilation parameters The MAP and HF were continuously monitored (Schmidt et al., 2015a). All

CE

administered fluids and blood samples were calculated to guarantee that all animals

AC

received an equal intravenous fluid replacement of 25ml/kg/h. A balanced solution (Sterofundin® ISO, B. Braun, Melsungen, Germany) was used as crystalloid baseline maintenance fluid

The animals were mechanically ventilated with a rodent ventilator (Harvard Apparatus Inspira Advanced Safety Volume Controlled Ventilator (55-7058, Harvard Apparatus, Massachusetts, USA; respiratory rate 70 breaths per min, tidal volume 7,5ml/kg, max. inspiratory pressure 10 mmH2O, inspiration time:expiration time 1:1.5,

8

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

FiO2 0,3-0,4). Adequate oxygenation and ventilation were regularly checked with blood gas analyses. The body core temperature was monitored with a rectal thermistor probe and maintained at 37± 1°C by placing the animals on heating pads. Maintenance fluid

PT

was indirectly warmed to 37°C by directing the catheter system through a warmed water bath and exposed intraabdominal tissues were continuously superperfused

RI

with 37°C crystalloid solution (Sterofundin® ISO, B. Braun, Melsungen, Germany).

SC

Chemicals and reagents

NU

A stock solution of LPS (Ultra pure lipopolysaccharide, Escherichia coli 0111:B4 strain- TLR4 ligand, InvivoGen, San Diego, CA 92121, USA) was prepared by

MA

dissolving LPS in saline to a concentration of 5 mg/ml. The solution was stored in a glass container at 5 °C. For the experiments, the stock solution was diluted in saline

D

to the appropriate concentration with an application volume of 0.5 – 1ml depending

PT E

on the experimental group. GTS-21 (3-(2,4-Dimethoxybenzylidene)-anabaseine dihydrochloride, ≥97% (HPLC), SML0326, Sigma-Aldrich Chemie GmbH, Steinheim,

CE

Germany) was administered in a dosage of 1mg/kg diluted in saline to the appropriate concentration with an application volume of 0.5 – 1ml depending on the

AC

experimental group. Also, 0.5 – 1ml crystalloid solution (Sterofundin® ISO, B. Braun, Melsungen, Germany) was administered depending on the experimental group. For measurement of erythrocyte velocity fluorescent-labeled erythrocytes from donor rats were injected 10 min before baseline measurements (0.5 mL/kg body weight; hematocrit 50%; labeled with a red fluorescent cell linker kit (PKH26-GL; Sigma Chemical, Deisenhofen, Germany). To quantify albumin leakage across mesenteric venules, 50 mg/kg of FITC-albumin (FITC-albumin, fluorescein isothiocyanate-

9

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

labeled bovine albumin; Sigma Chemicals, Deisenhofen, Germany) was injected 10 min before baseline measurements. Experimental design The experimental protocol was designed to evaluate the effect of GTS-21 in a co-,

PT

pre- and post-treatment setting in endotoxemic animals and to compare possible GTS-21 effects with an endotoxemic group 180 min after the baseline IVM (Fig. 1).

RI

The non-endotoxemic GTS-21 group was included in our experimental design to

SC

evaluate possible GTS-21 effects on the microcirculation independent of the

NU

endotoxemia setting. All animals were treated identically until completion of the baseline IVM. After completion of the baseline IVM the animals were randomized to

MA

the experimental groups [n=10/group] using the online Prism Graphpad® randomizer tool.

D

A 45 min stabilization period followed the exteriorization of the mesentery. At 10 min

PT E

before baseline IVM, 1.5 ml blood was taken from the left femoral vein and immediately replaced by an equal volume of fluorescent-labeled donor erythrocytes

CE

and FITC-albumin. The study design comprised two time points for the i.v. application of the test reagents (LPS, GTS-21, crystalloid solution). The first timepoint t0 was

AC

directly after the baseline IVM and t1 followed 60 min later upon completion of the second IVM. The test reagents were applied slowly via the left femoral vein over 5 min to guarantee hemodynamic stability. All animals received a total test reagent volume of 1 ml. Animals were grouped as follows: endotoxemia group (I) with LPS at t0; co-treatment group (II) with GTS-21 and LPS at t0; pre-treatment group (III) with GTS-21 at t0 and LPS at t1; post-treatment group (IV) with LPS at t0 and GTS-21 at t1; control group 10

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

(V) with crystalloid solution at t0; non-endotoxemic GTS-21 group (VI) with GTS-21 at t0 (Fig.1). The consecutive IVMs were performed at 60, 120 and 180 min after baseline IVM. The systemic inflammatory response was evaluated by measuring TNF-ɑ levels

PT

before baseline IVM and after the IVM at 180 min.

RI

In vivo intravital microscopy

SC

Intravital microscopy equipment, picture acquisition and evaluation criteria have been described in previous publications from our group (Schmidt et al., 2015a; Schmidt et

NU

al., 2015b). In short, intravital fluorescence microscopy was performed with a

MA

specially designed microscope (Orthoplan, Leica, Wetzlar, Germany) equipped with a 40-fold water immersion objective (Achroplan 40/0.75W; Zeiss, Jena, Germany). A digital camera (TypPS/DX4-285FW, Kappa opto-electronics GmbH, Gleichen,

PT E

D

Germany) equipped with a capturing software (Streampix 5.3.0, Norpix Inc., Montreal, Canada) was used for image recording, and image analysis was carried out with ImageJ (NIH, Bethesda, MD) and Histo (Version 3.0.2.4, Exp. Chirurgie,

CE

Uniklinik Heidelberg 2011). For the quantification of macromolecular leakage images

AC

were digitized and the gray levels reflecting fluorescent intensity were measured within the venule under study (iv) as well as in an equal and contiguous area of the perivenular interstitium (ii). Macromolecular leakage was determined as the ii/iv ratio (arbitrary units). Transillumination microscopy was used to evaluate leukocyteendothelial interactions. Adherent leukocytes were defined as cells that did not move or detach from the endothelial wall for 30 seconds and are expressed as cells/ 100 µm venule length. Venular wall shear rate was calculated by the Newtonian definition (γ =8 (VRBC/Dv)), using the measured vessel diameters and the mean red blood cell 11

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

velocities (VRBC) of 20–30 individual erythrocytes in single unbranched postcapillary venules. Measurement of TNF-α levels Blood samples were centrifuged at 2000 rpm for 10 min. Serum samples were frozen

PT

at -80°C. TNF-ɑ levels were quantified using a commercial ELISA kit according to the

RI

manufacturer instructions (Rat TNF-α ELISA MAXTM Deluxe Sets, BioLegend, San

SC

Diego, CA, USA).

NU

Statistical analysis

The IVM analysis contained all data, including outliers that fulfilled our inclusion

MA

criteria for postcapillary venules. The mean IVM data of 2-5 postcapillary venules were calculated as the mean value for one animal. The main outcome parameter for

D

this study was the perivenular macromolecular leakage around postcapillary venules.

PT E

The effectivity of LPS-induced microvascular inflammation with regard to perivenular macromolecular leakage and leukocyte adherence was assessed by the comparison

CE

of the endotoxemia group (I) with the control group (V). To evaluate the effect of GTS-21 on microvascular alterations during endotoxemia, endotoxemic GTS-21

AC

treated groups (II, IV) with an equal inflammatory hit (LPS 5mg/kg combined with a 180 min endotoxemia exposure time) are compared to the endotoxemia group (I) at the 180 min IVM. In a separate analysis, endotoxemia independent GTS-21 effects within our experimental model were evaluated by comparing the non-endotoxemic GTS-21 group (IV) with the control group (V) (presented in companion data in brief article). Results are expressed as median with interquartile range (Q1 – Q3) for each group unless otherwise noted. The D'Agostino-Pearson omnibus normality test was applied 12

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

to check for normal distribution. Due to non-normally distributed data, nonparametric methods for evaluation were used. For comparing two groups, a Mann-Whitney-U test was performed. In the case of more than two groups, a global Kruskal-Wallis test was done in advance, followed by Dunn’s multiple comparison test. For repeated

PT

measurements within groups, Friedman and Wilcoxon tests were done. Data were considered statistically significant at p < 0.05.

RI

All statistical analyses were performed using GraphPad Prism 7 for Mac OS X

SC

(GraphPad Software, La Jolla, CA, USA).

NU

Results

MA

General observations

All animals survived the experimental procedures with stable vital parameters over

D

the 240 min observation time, indicating robust experimental conditions and the

PT E

successful establishment of normotensive endotoxemia. No significant differences were observed between the experimental groups at baseline with regard to IVM

article).

CE

measurements and systemic TNF-α levels (presented in companion data in brief

AC

LPS increased microvascular permeability and systemic TNF-α levels Macromolecular leakage increased significantly in both the endotoxemia group (I) and the non-endotoxemic control group (V) after 180 min observation time compared to the baseline IVM (p< 0.05) (Fig 2A). Macromolecular leakage was significantly higher in the endotoxemia group (I) compared to the non-endotoxemic control group (V) at the 180 min IVM (p< 0.05) (Fig. 2B). No significant difference was observed between the endotoxemia group (I) and the non-endotoxemic control (V) regarding 13

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

adherent leukocytes (Fig. 2D) and venular wall shear rate (Fig. 2E) at the 180 min IVM. The TNF-α levels were significantly increased in the endotoxemia group (I) compared to the non-endotoxemic control (V) at the 180 min IVM (p< 0.05) (Fig. 2F). GTS-21 treatment ameliorates LPS-induced microvascular permeability

PT

Macromolecular leakage increased significantly in the endotoxemia group (I) and the

RI

endotoxemic co (II)- and post-treatment (IV) groups during the observation time

endotoxemic

animals

with

GTS-21

SC

compared to the respective baseline IVMs (p< 0.05) (Fig. 3A). Treatment of resulted

in

a

significant

reduction

of

MA

endotoxemia group (I) (p< 0.05) (Fig 3B).

NU

macromolecular leakage in the co(II)- and post-treatment (IV) group compared to the

Leukocyte-endothelial interactions and venular wall shear rate were not

D

affected by GTS-21 treatment in endotoxemic rats

PT E

The number of adherent leukocytes increased significantly within the endotoxemic groups (I, II and IV) compared to the respective baseline IVMs (p< 0.05) (Fig 4A).

CE

Treatment with GTS-21 did not result in a significant reduction of adherent leukocytes in endotoxemic animals (II, IV) compared to the endotoxemia group (I) after 180 min

AC

(Fig 4B). In line with the increased numbers of leukocytes within the endotoxemic groups (I, II, IV), venular wall shear rate was significantly reduced within the groups compared to their respective baseline IVMs (p< 0.05) (Figure 5A). No significant difference in the venular wall shear rate was observed between the endotoxemia group (I) and endotoxemic GTS-21 treated animals (II, IV) after 180 min (Fig 5B). Systemic TNF-α levels were not affected by GTS-21 treatment in endotoxemic rats

14

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

No significant differences in systemic TNF-ɑ levels were observed between the endotoxemic group I and the GTS21-treated endotoxemic groups II and IV (Fig 6). GTS-21 pre-treatment showed no effect on microcirculatory parameters The endotoxemic groups (I, II, IV) are characterized by an identical inflammatory hit

PT

with LPS 5mg/kg combined with an endotoxemia exposure time of 180 min after the baseline IVM. The pre-treatment setting in this study is methodically limited insofar

RI

that during the 180 min observation time with exteriorized mesentery and

SC

standardized IVMs, the endotoxemia exposure time is 120 min. After correction for

NU

the 120 min endotoxemia exposure time, there is no significant difference between the corresponding IVM results between the endotoxemia group (I) at the 120 min IVM

MA

and the pre-treatment group (III) at the 180 min IVM. Therefore GTS-21 pretreatment had no significant effect on macromolecular leakage, leukocyte adherence

D

and venular wall shear rate when corrected for the 120 min endotoxemia time in this

PT E

study (Fig 7).

Discussion

CE

The magnitude of microcirculatory dysregulation contributes to disease severity and progression in sepsis and septic shock. Here we report that intravenously

AC

administrated GTS-21 (1mg/kg) has a protective effect on microvascular permeability in endotoxemic animals when coadministered with LPS (co-treatment group II) or given one hour after established endotoxemia (post-treatment group IV). Leukocyte adhesion, venular wall shear rate and systemic TNF-ɑ levels were not affected in endotoxemic animals treated with GTS-21. Macromolecular leakage of FITC-labeled albumin was significantly increased in the endotoxemia group (I) compared to the non-endotoxemic control group (V), indicating disrupted endothelial integrity of the examined post capillary venules due to LPS-induced endothelial cell activation. The 15

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

effects of cholinergic α7nAChR agonists such as GTS-21 on microvascular alterations during inflammation are of interest for IVM evaluation, because vascular endothelial rat cells express the α7nAChR receptor (Moccia et al., 2004; Pena et al., 2011). The observed protective effect of GTS-21 on microvascular barrier function is

PT

in line with described beneficial effects of cholinergic agents on endothelial function during inflammation in vivo and in vitro (Chatterjee et al., 2009; Peter et al., 2010;

RI

Saeed et al., 2005; Schmidt et al., 2015a). Regarding sepsis-induced tissue edema

SC

formation, our results are in line with those of Nullens et al. who demonstrated that three repeated intraperitoneal injections of 8mg/kg of GTS-21 within 48 h decreased

NU

colonic permeability in a murine polymicrobial abdominal sepsis model (Nullens et al.,

MA

2016). GTS-21 administration protected these animals from developing septic ileus and caused a significant decrease in serum levels of IL-6 levels, whereas TNF-α

D

levels were not reduced (Nullens et al., 2016).

PT E

In line with previous publications (Peter et al., 2010; Schmidt et al., 2015a; Schmidt et al., 2015b), we observed a significant increase in leukocyte adherence combined with a significant reduction in wall shear rate within the endotoxemic groups (I, II, IV).

CE

However, the comparison of the endotoxemia group (I) with the control group (V)

AC

indicates reduced LPS effectivity in inducing distinctive leukocyte adherence in this study. Leukocyte-independent endothelial dysfunction has been demonstrated to be an important factor for microvascular damage in the early phase of endotoxemia (Walther et al., 2000). Therefore, the in vivo observations of this study should be evaluated in follow up studies with regard to the cellular functionality of leukocyte– endothelial interactions and specific intracellular endothelial mechanisms, because we do not provide causally determined agonist-receptor reactions of GTS-21 with α7nAChR receptors in mesenteric venular endothelial cells and on leukocytes. 16

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Nevertheless, our results point to a protective leukocyte-independent effect of GTS21 on endothelial barrier function, indicated by the significant reduction of macromolecular leakage in the co (II)- and the post-treatment (IV) groups. Giebelen et al. showed that 4 mg/kg GTS-21 i.p. inhibited the influx of neutrophils into the

PT

peritoneal cavity in endotoxemic mice. This inhibitory effect on neutrophil recruitment by GTS-21 was independent of its systemic effect on reducing TNF-α levels

RI

(Giebelen et al., 2007). Pavlov et al. showed that i.p. GTS-21 suppresses systemic

SC

TNF-α in endotoxemic mice and increased survival in endotoxemia in a dosedependent manner. Within the same study, this dose-dependent GTS-21 effect was

NU

also reproduced in a cecal ligation and puncture model showing increased anti-

MA

inflammatory efficacy and improved survival for higher GTS-21 doses (Pavlov et al., 2007). In this study, the observed beneficial effect on microvascular permeability is

D

contrasted by unaffected TNF-α levels in endotoxemic animals treated with GTS-21

PT E

(II, IV). This in vivo observation might be suggestive of a protective effect of GTS-21 on microvascular barrier function despite an ongoing LPS-induced systemic immune response. Given the established dose-dependent anti-inflammatory effect of GTS-21

CE

(Giebelen et al., 2007; Nullens et al., 2016; Pavlov et al., 2007), the single i.v.

AC

application of 1mg/kg GTS-21 might have been too low to affect leukocyteendothelial interactions and systemic TNF-α levels. Therefore, our results might reflect a differentiated dose-dependent GTS-21 effect on effector-systems within the inflammatory response. With regard to the established anti-inflammatory effect of preemptive administered GTS-21 in higher dosages (Giebelen et al., 2007; Nullens et al., 2016; Pavlov et al., 2007), the observed results for the pre-treatment group (III) could be explained by a dose-dependent GTS-21 effect and the methodological limitations addressed below. In contrast to the results of pre-treatment, the protective 17

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

effect on microvascular barrier function in the co- and post-treatment setting could point to a time-dependent anti-inflammatory GTS-21 effect relative to the occurrence of an inflammatory hit. Cai et al. showed that addition of GTS-21 to a resuscitation infusion protocol after established

severe

hemorrhage

prevented

systemic

PT

inflammation and improved survival dependent on dose in a rat model (Cai et al., 2009). Therefore, the observed in vivo effect of a single infusion of “low dose” GTS-

RI

21 given intravenously identifies GTS-21 as a substance with potential for treating

SC

sepsis-induced vascular leakage and tissue edema formation. Chatterjee et al. showed that IL-6-mediated endothelial cell activation was significantly attenuated by

NU

GTS-21 through the JAK2/STAT3 signaling pathway in macrovascular human

MA

umbilical vein endothelial cells and microvascular endothelial cells in vitro (Chatterjee et al., 2009). The stimulation of the α7nAChR by GTS-21 in human leukocytes

D

resulted in a dose-dependent inhibition of proinflammatory, but not anti-inflammatory

PT E

cytokine release in a TLR-dependent (Rosas-Ballina et al., 2009) and independent manner (Kox et al., 2009). GTS-21 underwent phase I/II trails in healthy subjects, rendering it suitable for human use and for the treatment of schizophrenia (Freedman

CE

et al., 2008; Kitagawa et al., 2003). In experimental human endotoxemia, oral GTS-

AC

21 administration showed large inter-individual plasma concentration variability with a correlation for increased anti-inflammatory effect with higher GTS-21 plasma concentrations (Kox et al., 2011a). Therefore, considering the protective effect of intravenous GTS-21 on microvascular permeability demonstrated in this study, GTS21 added to the resuscitation fluid regime could attenuate tissue edema formation without negatively compromising the immune response in ongoing human sepsis.

18

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Limitations We acknowledge that LPS endotoxemia is not equivalent to the complex clinical entity of sepsis and that the observed alterations in the mesenteric postcapillary venules may not reflect the microcirculatory situation in other vascular beds. The pre-

PT

treatment setting in our experimental design has the benefit of a standardized baseline IVM before administration of a test substance. Corrected for an effective 120

RI

min endotoxemia exposure time, pre-treatment with GTS-21 had no effect on

SC

microcirculatory parameters. Optimal stable experimental conditions limit the mesentery exteriorization time to 180 min in our experimental setting. These intrinsic

NU

methodological factors combined with a comparably low GTS-21 dosage could

MA

explain that the results of the pre-treatment group (III) are not in line with the beneficial pre-emptive anti-inflammatory effects of GTS-21 administrated in higher

D

dosages (Giebelen et al., 2007; Nullens et al., 2016; Pavlov et al., 2007). Preliminary

PT E

dose response studies showed that a single LPS dose of 5mg/kg induced sufficient microvascular leakage, in line with our previous studies with continuous LPS infusion (Peter et al., 2010; Schmidt et al., 2015a; Schmidt et al., 2015b). To mirror clinical

CE

practice and to provide efficient anesthesia combined with hemodynamic stability

AC

(Brammer et al., 1993) we introduced propofol/fentanyl anesthesia to our experimental setting. Propofol/fentanyl anesthesia has been demonstrated to suppress both rolling and adhesion of leukocytes in vivo in mesenterial postcapillary venules after experimental hemorrhage in a comparable experimental IVM setting (Brookes et al., 2006). To guarantee catheter patency during the observation time, all intravascular catheters had to be flushed once with heparin (50 IU) before implantation. We acknowledge that the combination of the above mentioned methodological factors could have influenced leukocyte adherence and venular wall 19

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

shear rate, resulting in a non-significant difference between endotoxemic and nonendotoxemic animals when compared to our previous studies (Peter et al., 2010; Schmidt et al., 2015a). GTS-21 has a high affinity for the rodent α7nAChR but its affinity for human α7nAChR is described to be low. Its metabolite, 4OH-(GTS-21) has

PT

been described to be a selective partial agonist with greater efficacy for both human and rat α7nAChRs (Conejero-Goldberg et al., 2008; Kem, 2000). Moreover GTS-21

RI

is not solely specific for α7nAChR, it also affects other receptors including α4β2-

SC

nAChRs and 5-HT3AR (Conejero-Goldberg et al., 2008). Leukocyte-independent plasma extravasation during early endotoxemia can be inhibited by serotonin (5-HT)

NU

receptor antagonists (Walther et al., 2002; Walther et al., 2001). Underlying

MA

α7nAChR mechanisms regarding endothelial permeability and leukocyte–endothelial interactions, as well as crosslinks to serotonin mediated effects, need to be further

D

evaluated in in vitro studies.

PT E

Conclusion

Considering its anti-inflammatory efficacy and safety profile (Conejero-Goldberg et al., 2008; Kem, 2000), GTS-21 is a substance with potential for clinical treatment. A

CE

single i.v. infusion of “low dose” (1mg/kg) GTS-21 proves effective in reducing LPS-

AC

induced microvascular permeability when coadministered directly or given one hour after endotoxemia induction in rats. We suggest that these data may be used as a basis for an approach to restore microvascular function in human sepsis.

Acknowledgments The authors would like to thank Roland Galmbacher and Ute Krauser for outstanding technical support, and Anuradha Gunale MD, PhD for revising the manuscript critically. 20

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

References

AC

CE

PT E

D

MA

NU

SC

RI

PT

Brammer, A., et al., 1993. A comparison of propofol with other injectable anaesthetics in a rat model for measuring cardiovascular parameters. Lab Anim. 27, 250-7. Brookes, Z. L., et al., 2006. Intravenous anesthesia inhibits leukocyte-endothelial interactions and expression of CD11b after hemorrhage. Shock. 25, 492-9. Cai, B., et al., 2009. Alpha7 cholinergic-agonist prevents systemic inflammation and improves survival during resuscitation. J Cell Mol Med. 13, 3774-85. Chatterjee, P. K., et al., 2009. Cholinergic agonists regulate JAK2/STAT3 signaling to suppress endothelial cell activation. Am J Physiol Cell Physiol. 297, C1294-306. Conejero-Goldberg, C., et al., 2008. Alpha7 nicotinic acetylcholine receptor: a link between inflammation and neurodegeneration. Neurosci Biobehav Rev. 32, 693-706. De Backer, D., et al., 2014. Pathophysiology of microcirculatory dysfunction and the pathogenesis of septic shock. Virulence. 5, 73-9. de Jonge, W. J., Ulloa, L., 2007. The alpha7 nicotinic acetylcholine receptor as a pharmacological target for inflammation. Br J Pharmacol. 151, 915-29. Freedman, R., et al., 2008. Initial phase 2 trial of a nicotinic agonist in schizophrenia. Am J Psychiatry. 165, 1040-7. Giebelen, I. A., et al., 2007. Stimulation of alpha 7 cholinergic receptors inhibits lipopolysaccharideinduced neutrophil recruitment by a tumor necrosis factor alpha-independent mechanism. Shock. 27, 443-7. Kem, W. R., 2000. The brain alpha7 nicotinic receptor may be an important therapeutic target for the treatment of Alzheimer's disease: studies with DMXBA (GTS-21). Behav Brain Res. 113, 16981. Khan, M. A., et al., 2012. Lipopolysaccharide upregulates alpha7 acetylcholine receptors: stimulation with GTS-21 mitigates growth arrest of macrophages and improves survival in burned mice. Shock. 38, 213-9. Kirkpatrick, C. J., et al., 2001. The non-neuronal cholinergic system in the endothelium: evidence and possible pathobiological significance. Jpn J Pharmacol. 85, 24-8. Kitagawa, H., et al., 2003. Safety, pharmacokinetics, and effects on cognitive function of multiple doses of GTS-21 in healthy, male volunteers. Neuropsychopharmacology. 28, 542-51. Kox, M., et al., 2011a. Effects of the alpha7 nicotinic acetylcholine receptor agonist GTS-21 on the innate immune response in humans. Shock. 36, 5-11. Kox, M., et al., 2011b. alpha7 nicotinic acetylcholine receptor agonist GTS-21 attenuates ventilatorinduced tumour necrosis factor-alpha production and lung injury. Br J Anaesth. 107, 559-66. Kox, M., et al., 2009. GTS-21 inhibits pro-inflammatory cytokine release independent of the Toll-like receptor stimulated via a transcriptional mechanism involving JAK2 activation. Biochem Pharmacol. 78, 863-72. Moccia, F., et al., 2004. Expression and function of neuronal nicotinic ACh receptors in rat microvascular endothelial cells. Am J Physiol Heart Circ Physiol. 286, H486-91. Nullens, S., et al., 2016. Effect of Gts-21, an Alpha7 Nicotinic Acetylcholine Receptor Agonist, on ClpInduced Inflammatory, Gastrointestinal Motility, and Colonic Permeability Changes in Mice. Shock. 45, 450-9. Pavlov, V. A., et al., 2007. Selective alpha7-nicotinic acetylcholine receptor agonist GTS-21 improves survival in murine endotoxemia and severe sepsis. Crit Care Med. 35, 1139-44. Pena, V. B., et al., 2011. alpha 7-type acetylcholine receptor localization and its modulation by nicotine and cholesterol in vascular endothelial cells. J Cell Biochem. 112, 3276-88.

21

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

SC

RI

PT

Peter, C., et al., 2010. Effects of physostigmine on microcirculatory alterations during experimental endotoxemia. Shock. 33, 405-11. Pohanka, M., 2012. Alpha7 nicotinic acetylcholine receptor is a target in pharmacology and toxicology. Int J Mol Sci. 13, 2219-38. Rosas-Ballina, M., et al., 2009. The selective alpha7 agonist GTS-21 attenuates cytokine production in human whole blood and human monocytes activated by ligands for TLR2, TLR3, TLR4, TLR9, and RAGE. Mol Med. 15, 195-202. Saeed, R. W., et al., 2005. Cholinergic stimulation blocks endothelial cell activation and leukocyte recruitment during inflammation. J Exp Med. 201, 1113-23. Schmidt, K., et al., 2015a. Cytidine-5-diphosphocholine reduces microvascular permeability during experimental endotoxemia. BMC Anesthesiol. 15, 114. Schmidt, K., et al., 2015b. Time-dependent effect of clonidine on microvascular permeability during endotoxemia. Microvasc Res. 101, 111-7. Tracey, K. J., 2007. Physiology and immunology of the cholinergic antiinflammatory pathway. J Clin Invest. 117, 289-96. Tracey, K. J., 2009. Reflex control of immunity. Nat Rev Immunol. 9, 418-28. Walther, A., et al., 2002. Selective serotonin receptor antagonism and leukocyte-independent plasma extravasation during endotoxemia. Microvasc Res. 63, 135-8. Walther, A., et al., 2000. Leukocyte-independent plasma extravasation during endotoxemia. Crit Care Med. 28, 2943-8. Walther, A., et al., 2001. Methysergide attenuates leukocyte-independent plasma extravasation during endotoxemia. J Crit Care. 16, 121-6.

22

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Figure legends

PT

Fig. 1. Experimental protocol

Following a stabilization period after surgical preparation intravital microscopic measurements (IVM)

RI

were performed at 60, 120 and 180 min after baseline IVM in endotoxemic and non-endotoxemic

SC

animals (n=10/group). The study design comprised intravenous infusion of LPS 5mg/kg, GTS-21 1mg/kg and crystalloid solution in a total test reagent volume of 1ml per animal over 5 min. Application

NU

time points: t0 directly after baseline IVM and t1 followed 60 min later upon completion of the second IVM. All administered fluids were calculated to guarantee that all animals received equal amounts of

MA

intravenous fluids.

PT E

D

Fig. 2. LPS increases microvascular permeability and systemic TNF levels.

Fig 2A: Change of macromolecular leakage during the observation time in the endotoxemia group (I)

CE

and the control group (V) illustrating effective LPS-induced microvascular leakage. In both groups, macromolecular leakage increased significantly after 180 min compared to the respective baseline

AC

IVM (#180 min IVM vs baseline IVM: p< 0.05; macromolecular leakage is expressed as the ratio of perivenular to venular fluorescence intensity in arbitrary units, medians and interquartile range (Q1– Q3) are displayed).

Fig 2B: Comparison between the endotoxemia (I) and the control (V) group 180 min after baseline IVM. LPS induced significant macromolecular leakage, indicative of a disrupted endothelial integrity in the examined post capillary venules due to microvascular inflammation (*LPS vs. control; p< 0.05; scatter plots with medians and interquartile range (Q1–Q3) are displayed).

23

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Fig 2C: Exemplary IVM images demonstrating the LPS effect on microvascular permeability during endotoxemia. Fluorescent IVM images show postcapillary venules recorded at baseline IVM (upper panels) and 180 min later (lower panels). Fig 2D: The increase in the number of adherent leukocytes in the endotoxemia group (I) was not significant (ns) compared to the control group (V) 180 min after baseline IVM (Adherent leukocytes are

PT

expressed as cells/ 100 µm venule length, scatter plots with medians and interquartile range (Q1–Q3)

RI

are displayed).

Fig.2E: The reduction in venular wall shear rate in postcapillary venules in the endotoxemia group (I)

SC

was not significant (ns) compared to the control group (V) 180 min after baseline IVM. (Venular wall -1

shear rate is expressed as s , scatter plots with medians and interquartile range (Q1–Q3) are

NU

displayed).

Fig 2F: The TNF-α levels were significantly increased after 180 min in the endotoxemia group (I)

MA

compared to the control group (VI) (*LPS vs. control; p< 0.05; TNF α levels are expressed as pg/ml,

PT E

D

scatter plots with medians and interquartile range (Q1–Q3) are displayed).

CE

Fig. 3. Effect of GTS-21 administration on macromolecular leakage

Fig 3A: Change in macromolecular leakage during the observation time after baseline IVM in the

AC

endotoxemia (I) and GTS-21 treated endotoxemic groups (II, IV). Macromolecular leakage increased significantly in all endotoxemic groups after 180 min compared to the respective baseline IVM (#180 min IVM vs baseline IVM: p< 0.05; macromolecular leakage is expressed as the ratio of perivenular to venular fluorescence intensity in arbitrary units). Fig 3B: Intravenous application of 1mg/kg GTS-21 simultaneously with (co-treatment group (II)) and 1h after (post-treatment group (IV)) endotoxemia induction reduced macromolecular leakage significantly compared to the endotoxemia group (I) after 180 min (*LPS vs. GTS-21 co-treatment; p< 0.05, *LPS vs. GTS-21 post-treatment; p< 0.05).

24

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

RI

PT

Fig. 4. Effect of GTS-21 administration on leukocyte adherence

Fig 4A: Change in leukocyte adherence during the observation time in the endotoxemia (I) and GTS-

SC

21 treated endotoxemic groups (II, IV). In all endotoxemic groups leukocyte adherence increased significantly after 180 min compared to the respective baseline IVM (#180 min IVM vs baseline IVM:

NU

p< 0.05; adherent leukocytes are expressed as cells/ 100 µm venule length)

MA

Fig 4B: GTS-21 co-treatment (II) and post-treatment (IV) had no significant effect on the number of adhering leukocytes during endotoxemia 180 min after baseline IVM compared to the endotoxemia (I)

PT E

D

group (scatter plots with medians and interquartile range (Q1–Q3) are displayed).

CE

Fig. 5. Effect of GTS-21 administration on venular wall shear rate

Fig 5A: Change in venular wall shear rate during the observation time in the endotoxemia (I) and GTS-

AC

21 treated endotoxemic groups (II, IV). In all endotoxemic groups venular wall shear rate was significantly reduced after 180 min compared to the respective baseline IVM (#180 min IVM vs -1

baseline IVM; venular wall shear rate is expressed as s ) Fig 5B: GTS-21 co-treatment (II) and post-treatment (IV) had no significant effect on venular wall shear rate compared with the endotoxemia group (I) 180 min after baseline IVM (scatter plots with medians and interquartile range (Q1–Q3) are displayed).

25

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Fig. 6. Systemic TNF-alpha levels were not affected by GTS-21 treatment in endotoxemic rats

Fig 6: Exemplary fluorescent IVM images of postcapillary venules with approximate median macromolecular leakage values of the endotoxemia- (I), the GTS-21 co- treatment (II) and the post-

PT

treatment groups (IV) at the 180 min IVM are presented in the upper panel. These IVM pictures display reduced macromolecular leakage in the GTS-21 treated endotoxemic groups (II, IV) compared

RI

to the endotoxemia group (I). The lower panel shows that there is no significant (ns) difference

SC

between the TNF α levels of the endotoxemia(I)-, the GTS-21 co-treatment(II)- and post-treatment (IV) groups. These results might suggest a beneficial local microvascular effect of GTS-21 treatment

NU

despite an ongoing systemic pro-inflammatory reaction. (TNF α levels are expressed as pg/ml; scatter

MA

plots with medians and interquartile range (Q1–Q3) are displayed).

D

Fig. 7. GTS-21 pre-treatment showed no effect on microcirculatory parameters

PT E

after correction for 120 min endotoxemia exposure time

CE

Fig 7. shows from left to right the comparisons between the endotoxemia group (I) at the 120 min IVM and the GTS-21 pre-treatment group (II) at the 180 min IVM for macromolecular leakage, leukocyte-

AC

adherence and venular wall shear rate (scatter plots with medians and interquartile range (Q1–Q3) are displayed). After correcting the analysis for a comparable endotoxemia exposure time of 120 min no significant effects between the LPS alone group (I) at the 120 min IVM and the GTS-21 pre-treatment group (II) at the 180 min IVM were observed.

26

ACCEPTED MANUSCRIPT

SC

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

Fig. 1

27

ACCEPTED MANUSCRIPT

SC

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

Fig. 2

28

ACCEPTED MANUSCRIPT

SC

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

Fig. 3

29

ACCEPTED MANUSCRIPT

SC

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

Fig. 4

30

ACCEPTED MANUSCRIPT

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

SC

Fig. 5

31

ACCEPTED MANUSCRIPT

SC

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

Fig. 6

32

ACCEPTED MANUSCRIPT

RI

PT

GTS-21 reduces microvascular permeability during experimental endotoxemia

AC

CE

PT E

D

MA

NU

SC

Fig. 7

33

ACCEPTED MANUSCRIPT GTS-21 reduces microvascular permeability during experimental endotoxemia

Highlights

PT

Intravital microscopy study GTS-21 has a protective effect on microvascular permeability when it is

CE

PT E

D

MA

NU

SC

RI

applied simultaneously or 1 hour after endotoxemia induction

AC

 

34