ORIGINAL ARTICLES
Treatment With a Novel Hemigramicidin-TEMPO Conjugate Prolongs Survival in a Rat Model of Lethal Hemorrhagic Shock Carlos A. Macias, MD,* Jeffrey W. Chiao, BS,* Jingbo Xiao, PhD,† Devinder S. Arora, PhD,* Yulia Y. Tyurina, PhD,‡ Russell L. Delude, PhD,* Peter Wipf, PhD,† Valerian E. Kagan, PhD,‡ and Mitchell P. Fink, MD*§
Objective: We sought to develop a therapeutic agent that would permit prolongation of survival in rats subjected to lethal hemorrhagic shock (HS), even in the absence of resuscitation with asanguinous fluids or blood. Methods and Results: We synthesized a series of compounds that consist of the electron scavenger and superoxide dismutase mimic, 4-amino-2,2,6,6-tetramethylpiperidine-N-oxyl (4-NH2-TEMPO), conjugated to fragments and analogs of the membrane-active cyclopeptide antibiotic, gramicidin S. Using an in vivo assay, wherein isolated intestinal segments were loaded inside the lumen with various test compounds, we studied these compounds for their ability to prevent ileal mucosal barrier dysfunction induced by subjecting rats to profound HS for 2 hours. The most active compound in this assay, XJB-5-131, ameliorated peroxidation of the mitochondrial phospholipid, cardiolipin, in ileal mucosal samples from rats subjected to HS. XJB-5-131 also ameliorated HS-induced activation of the pro-apoptotic enzymes, caspases 3 and 7, in ileal mucosa. Intravenous treatment with XJB-5-131 (2 mol/kg) significantly prolonged the survival of rats subjected to profound blood loss (33.5 mL/kg) despite administration of only a minimal volume of crystalloid solution (2.8 mL/kg) and the absence of blood transfusion. Conclusion: These data support the view that mitochondrially targeted electron acceptors and SOD mimics are potentially valuable therapeutics for the treatment of serious acute conditions, such as HS, which are associated with marked tissue ischemia. (Ann Surg 2007;245: 305–314)
From the Departments of *Critical Care Medicine, †Chemistry, ‡Environmental and Occupational Health, and §Surgery, University of Pittsburgh, Pittsburgh, PA. Supported by the Defense Advanced Research Projects Administration (DARPA contract W81XWH-05-2-0026) as well as U.S. Public Health Service National Institutes of Health (GM067082). Reprints: Mitchell P. Fink, MD, 616A Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261. E-mail:
[email protected]. Copyright © 2007 by Lippincott Williams & Wilkins ISSN: 0003-4932/07/24502-0305 DOI: 10.1097/01.sla.0000236626.57752.8e
Annals of Surgery • Volume 245, Number 2, February 2007
T
rauma ranks fifth as a cause of death among people of all ages living in the United States, and it is the leading cause of death among people younger than 45 years.1 In the United States, traumatic injuries result in approximately 100,000 deaths per year.1 Early deaths due to trauma are secondary to exsanguination or overwhelming central nervous system injuries, whereas late deaths are secondary to sepsis and multiple organ system dysfunction syndrome.2,3 Severe hypovolemia from hemorrhage is a major causative factor in almost half of these deaths, especially during the acute period (⬍2 hours) after injury.2,3 Currently, the primary strategy for treating hemorrhagic shock (HS) is to control ongoing bleeding and restore intravascular volume by infusing an asanguinous fluid and packed red blood cells. If intravascular volume expansion successfully restores cardiac output and arterial blood pressure before definitive hemostasis has been achieved, then, paradoxically, resuscitation can promote bleeding and shorten survival.4 – 6 Additionally, conventional approaches toward resuscitation require administration of large volumes of fluids, which are intrinsically heavy and bulky. Thus, logistic considerations can sometimes limit the capacity of first responders to provide adequate conventional resuscitation. In view of these considerations, it would be desirable to be able to administer a therapeutic agent in the field to patients with profound HS so that survival could be prolonged until it is feasible to obtain surgical control of bleeding vessels. Ideally, administration of this agent would prolong survival (ie, extend the “golden hour”) without increasing blood pressure to such an extent that endogenous hemostatic mechanisms are disrupted. During HS, microvascular perfusion to tissues is compromised, leading to cellular hypoxia. Under hypoxic conditions, mitochondria leak electrons, leading to the formation of partially reduced forms of molecular oxygen, particularly the free radical, superoxide anion (O2·⫺).7–9 Accordingly, we hypothesized that treatment with a potent pharmacologic agent that is capable of scavenging O2·⫺ produced by mitochondria might prolong survival during HS. Stable nitroxide radicals seem to be ideal candidates for this purpose as they can act as one-electron acceptors (competing with molecular oxygen) to yield hydroxylamines. The latter are potent radical
305
Annals of Surgery • Volume 245, Number 2, February 2007
Macias et al
scavengers producing nitroxide radicals upon their reaction with reactive species. In this way, nitroxide radicals can be recycled and provide effective protection against O2·⫺ generation.10 –12 An important aspect of this design was the use of a targeting sequence that delivered the scavenging agent preferentially to the mitochondrial membrane. In accordance with this concept, we synthesized a series of compounds that consist of 4-amino-2,2,6,6-tetramethylpiperidine-N-oxyl (4-NH2-TEMPO) conjugated to fragments of the membrane-active cyclopeptide antibiotic, gramicidin S. The gramicidin segments were used to target the TEMPO “payload” to mitochondria because antibiotics of this group have a high affinity to bacterial membranes13 and because of the close relationship between bacteria and mitochondria. In the present study, we used an in vivo screening assay to evaluate the pharmacologic effects of several related gramicidin S-based compounds. All compounds incorporated the hemigramicidin motif or the corresponding (E)-alkene peptide bond isosteres. In some of the compounds, the “targeting component” was covalently linked to 4-NH2-TEMPO, whereas in others the hemigramicidin motif was capped with a simple ester. Herein, we show that treatment with a very small volume of a solution of the most active compound, XJB-5-131, significantly prolonged the survival of rats subjected to massive blood loss, even though the animals were not resuscitated with either blood or other fluids and remained profoundly hypotensive.
MATERIALS AND METHODS Materials All chemicals were from Sigma-Aldrich (St. Louis, MO) unless otherwise noted. The hemigramicidin-based compounds were synthesized as previously described.14 Heparin, ketamine HCl, and sodium pentobarbital were from Abbott Laboratories (North Chicago, IL). Dulbecco modified Eagle medium (DMEM) was from BioWhittaker (Walkersville, MD). Fetal bovine serum (FBS; ⬍0.05 endotoxin units/mL) was from Hyclone (Logan, UT). Pyrogen-free sterile normal saline solution was from Baxter (Deerfield, IL).
Cells Caco-2BBe human enterocyte-like epithelial cells were obtained from the American Type Culture Collection (Manassas, VA). Cells were routinely maintained at 37°C under a humidified atmosphere containing 8% CO2 in air. The culture medium was DMEM supplemented with 10% FBS, nonessential amino acids supplement (Sigma-Aldrich catalogue # M7145), sodium pyruvate (2 mmol/L), streptomycin (0.1 mg/mL), penicillin G (100 U/mL), and human transferrin (0.01 mg/mL). The culture medium was changed 3 times per week.
Surgical Procedures to Obtain Vascular Access All study protocols using rats followed the guidelines for the use of experimental animals of the U.S. National Institutes of Health and were approved by the Institutional Animal Care and Use Committee at the University of Pitts-
306
burgh. Male specific pathogen-free Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA), weighing 150 to 250 g, were housed in a temperature-controlled environment with a 12-hour light/dark cycle. The rats had free access to food and water. For experiments, rats were anesthetized with intramuscular ketamine HCl (30 mg/kg) and intraperitoneal sodium pentobarbital (35 mg/kg). Animals were kept in a supine position during the experiments. Lidocaine (0.5 mL of a 0.5% solution) was injected subcutaneously to provide local anesthesia at surgical cutdown sites. To secure the airway, a tracheotomy was performed and polyethylene tubing (PE 240; Becton Dickinson, Sparks, MD) was introduced into the trachea. Animals were allowed to breathe spontaneously. The right femoral artery was cannulated with polyethylene tubing (PE 10). This catheter was attached to a pressure transducer that allowed instantaneous measurement of mean arterial pressure (MAP) during the experiment. For experiments using the pressure-controlled HS model, the right jugular vein was exposed, ligated distally, and cannulated with polyethylene tubing (PE 10) to withdraw blood. For experiments using the volume-controlled HS model, the jugular catheter was used to infuse the resuscitation solution and the right femoral vein, which was cannulated with a silicon catheter (Chronic-Cath, Norfolk Medical, Skokie, IL), was used to withdraw blood. All animals were instrumented within 30 minutes. Heparin (500 U/kg) was administered immediately after instrumentation through the femoral vein. Animals were placed on a thermal blanket to maintain their body temperature at 37°C. The positioning of the different devices aforementioned was checked postmortem.
Intestinal Mucosal Permeability Assay We sought to develop an in vivo assay system that would avoid the necessity to administer the compounds systemically to experimental animals, but instead would permit assessments of the test compounds in a local milieu. Recognizing that HS in rats leads to marked derangements in intestinal mucosal barrier function,15,16 we used ileal mucosal permeability to FD4 as a readout. For the assay, animals were allowed access to water but not food for 24 hours prior to the experiment to decrease the volume of intestinal contents. The rats were instrumented as described above. A midline laparotomy was performed, and the small intestine was exteriorized from the duodenojejunal junction to the ileocecal valve. A small incision was made on the antimesenteric aspect of the proximal small intestine and saline solution (1.5 mL) was injected. The bowel was ligated proximally and distally to the incision with 4-0 silk. The small intestine was compressed gently in aboral direction along its length to displace intestinal contents into the colon. Starting 5 cm from the ileocecal valve, the ileum was partitioned into 6 contiguous water-tight segments. Each segment was 3 cm long and was bounded proximally and distally by constricting circumferential 4-0 silk sutures. Care was taken to ensure that the vascular supply to the intestine was not compromised such that each segment was well perfused. Two randomly selected segments in each rat were injected with 0.3 mL of vehicle and served as internal controls to account for animal-to-animal variations in © 2007 Lippincott Williams & Wilkins
Annals of Surgery • Volume 245, Number 2, February 2007
the severity of shock or the response of the mucosa to it. To fill the segments, a small incision was made and the solution was injected using a Teflon catheter (Abbocath 16Ga, Abbot Laboratories). The remaining 4 other segments were injected with solutions containing either 4-hydroxy-2,2,6,6tetramethylpiperidine-N-oxyl (TEMPOL) or one of the gramicidin S-based compounds. Four different final concentrations of TEMPOL in normal saline were evaluated: 0.1, 1, 5, and 20 mmol/L. The gramicidin S-based compounds were dissolved in a mixture of DMSO and normal saline (1:99 vol/vol) and injected at final concentrations of 0.1, 1, 10, or 100 mol/L. After the segments were loaded with saline or the test compounds, the bowel was replaced inside the peritoneal cavity and the abdominal incision was temporarily closed using Backhaus forceps. After a 5-minute stabilization period, HS was induced by withdrawing blood via the jugular catheter. MAP was maintained at 30 ⫾ 3 mm Hg for 2 hours. The shed blood was reinfused as needed to maintain MAP within the desired range. After 2 hours of shock, the animals were euthanized with an intracardiac KCl bolus injection. The ileum was rapidly excised from the ileocecal valve to the most proximal gut segment. The tips of each segment were discarded. The small intestine was opened along the antimesenteric border; the mucosa was scraped away with the use of a glass microscope slide and immediately snap frozen in liquid nitrogen. The samples were stored at ⫺80°C until assayed for caspase 3/7 activity and peroxidation of phospholipids. For permeability measurements, each segment was converted into an everted gut sac, as previously described by Wattanasirichaigoon et al.16 Briefly, the sacs were prepared in ice-cold modified Krebs-Henseleit bicarbonate buffer (KHBB, pH 7.4). One end of the gut segment was ligated with a 4-0 silk suture; the segment was then everted onto a thin plastic rod. The resulting gut sac was mounted on a Teflon catheter connected to a 3-mL plastic syringe containing 1.5 mL of KHBB. The sac was suspended in a beaker containing KHBB plus FITC-labeled dextran (average molecular mass 4 kDa; FD4; 0.1 mg/mL). This solution was maintained at 37°C, and oxygenated by bubbling with a gas mixture (O2 95%/CO2 5%). After 30 minutes, the fluid within the gut sac was collected. The samples were cleared by centrifugation at 2000g for 5 minutes. Fluorescence of FD4 in the solution inside the beaker and within each gut sac was measured using a fluorescence spectrophotometer (LS-50, Perkin-Elmer, Palo Alto, CA) at an excitation wavelength of 492 nm and an emission wavelength of 515 nm. Mucosal permeability was expressed as a clearance normalized by the length of the gut sac with units of nL · min⫺1 · cm⫺1, as previously described.17 Results for a specific experimental condition (ie, specific test compound at a single concentration) were expressed as relative change in permeability calculated according to this equation: Relative change in permeability (%) ⫽ (CHS exp ⫺ Cnormal)/(CHS cont ⫺ Cnormal) ⫻ 100, where CHS exp is the clearance of FD4 measured for a gut segment loaded with the experimental compound, Cnormal is the clearance of FD4 measured in 6 gut segments from 3 normal animals not © 2007 Lippincott Williams & Wilkins
Hemigramicidin-TEMPO Conjugate
subjected to HS, CHS cont is the mean clearance of FD4 measured in 2 gut segments filled with vehicle from the same animal used to measure CHS exp.
Measurement of Permeability of Caco-2BBe Monolayers Monolayers of the enterocyte-like cell line, Caco-2, have been used extensively by our laboratory18,19 as well as others20,21 to study the physiology and pathophysiology of intestinal barrier function. The permeability of Caco-2 monolayers increases when the cells are incubated with the ROS, hydrogen peroxide (H2O2),21 or menadione, a redox-cycling quinone that promotes the formation of O2·⫺.20 Caco-2BBe cells (passages 15–22) were plated at a density of 5 ⫻ 104 cells/well on permeable filters (0.4 m pore size) in 12-well bicameral chambers (Transwell, Costar, Corning, NY). After 21 to 24 days, paracellular permeability was determined by measuring the apical-to-basolateral clearance of FD4 as previously described.22 Briefly, the medium on the basolateral side was replaced with control medium or medium containing menadione (50 mol/L final). Medium containing FD4 (25 mg/mL) was applied to the apical chamber. In some cases, one of the gramicidin S-based compounds, XJB-5-131, also was added to the apical side at final concentrations of 0.1, 1, 10, or 100 mol/L. After 6 hours of incubation, the medium was aspirated from both compartments. Permeability of the monolayers was expressed as a clearance (pL · h⫺1 · cm⫺2), as previously described.22
Caspase 3/7 Activity Assay Caspase 3/7 activity was measured using a commercially available assay kit, Caspase GloTM 3/7 assay kit (Promega, Madison, WI). Briefly, 50 L of rat gut mucosa homogenate (20 g protein) was mixed with 50 L of Caspase-GloTM reagent and incubated at room temperature for 1 hour. At the end of incubation period, the luminescence of each sample was measured using a plate reading chemiluminometer (ML1000, Dynatech Laboratories, Horsham, PA). Activity of caspase-3/7 was expressed as luminescence intensity (arbitrary units per mg protein). Protein concentrations were determined using the BioRad assay (Bio-Rad Laboratories, Inc., Hercules, CA).
Assay for Peroxidation of Phospholipids Gut mucosal samples were homogenized. Lipids were extracted from homogenates using the Folch procedure23 and resolved by 2D HPTLC as previously described.24 Spots of phospholipids were scraped from HPTLC plates and phospholipids were extracted from silica. Lipid phosphorus was determined by a micromethod.25 Oxidized phospholipids were hydrolyzed by pancreatic phospholipase A2 (2 U/L) in 25 mmol/L phosphate buffer containing 1 mmol/L CaCl2, 0.5 mmol/L EDTA, and 0.5 mmol/L SDS (pH 8.0, at room temperature for 30 minutes). Fatty acid hydroperoxides formed were determined by fluorescence HPLC of resorufin stoichiometrically formed during their microperoxidase 11catalized reduction in presence of Amplex Red (for 40 minutes at 4°C).26 Fluorescence HPLC (Eclipse XDB-C18 column, 5 m, 150 ⫻ 4.6 mm, mobile phase was composed of
307
Macias et al
25 mmol/L disodium phosphate buffer (pH 7.0)/methanol (60:40 vol/vol); excitation wavelength 560 nm, emission wavelength 590 nm) was performed on a Shimadzu LC100AT HPLC system equipped with fluorescence detector (RF-10Axl) and autosampler (SIL-10AD).
Survival of Rats Subjected to Volume-Controlled HS Following surgical preparation and a 5-minute stabilization period to obtain baseline readings, rats were subjected to HS. Bleeding was carried out in 2 phases. Initially, 21 mL/kg of blood was withdrawn over 20 minutes. Immediately thereafter, an additional 12.5 mL/kg of blood was withdrawn over 40 minutes. Thus, hemorrhage occurred over a total period of 60 minutes and the total blood loss was 33.5 mL/kg or approximately 55% of total blood volume. Rats were randomly assigned to receive XJB-5-131 (2 mol/kg) or its vehicle, a 33:67 (vol/vol) mixture of DMSO and normal saline. XJB-5-131 solution or vehicle alone was administered as a continuous infusion during the last 20 minutes of the hemorrhage period. The total volume of fluid infused was 2.8 mL/kg, and it was administered intravenously using a syringe pump (KD100, KD Scientific, New Hope, PA). Rats were observed for 6 hours or until expiration (defined by apnea for ⬎1 minute). At the end of the 6-hour observation period, animals that were still alive were euthanized with an overdose of KCl. Blood pressure was recorded continuously using a commercial strain-gauge transducer, amplifier, and monitor (S90603a, SpaceLabs, Redmond, WA). Blood samples (0.5 mL) were collected from the jugular vein at the beginning of hemorrhage (baseline), at the end of hemorrhage (shock), and at the end of resuscitation (resuscitation). We determined hemoglobin concentration 关Hb兴, lactate and glucose concentration using an auto-analyzer (Model ABL 725, Radiometer Copenhagen, Westlake, OH).
Data Presentation and Statistics
All variables are presented as mean ⫾ SEM. Statistical significance of differences among groups were determined
Annals of Surgery • Volume 245, Number 2, February 2007
using ANOVA and LSD tests, or Kruskal-Wallis and MannWhitney U tests as appropriate. Survival data were analyzed using the log-rank test. Significance was declared for P values less than 0.05.
RESULTS In Vivo Intestinal Mucosa Permeability Using the gut sac approach described in Methods, we evaluated 8 compounds: TEMPOL, one dipeptidic TEMPO analog, 3 hemigramicidin-TEMPO conjugates, and 3 hemigramicidin compounds lacking the TEMPO “payload.” The structures of these compounds are depicted in Figure 1. As expected, the mucosal permeability of intestinal segments from hemorrhaged animals was significantly greater than the permeability of segments from normal rats (52.3 ⫾ 0.5 vs. 6.9 ⫾ 0.1 nL · min⫺1 · cm2, respectively; P ⬍ 0.01). TEMPOL was previously shown to ameliorate organ damage27,28 when used to treat rats subjected to HS. Accordingly, we used intraluminal TEMPOL as a “positive control” for the gut mucosal protection assay. As anticipated, TEMPOL concentrations ⱖ1 mmol/L in the gut lumen ameliorated HS-induced ileal mucosal hyperpermeability (Fig. 2A). Two of the TEMPO conjugates, namely, XJB-5-208 (Fig. 2B) and XJB5-131 (Fig. 2E), also significantly ameliorated HS-induced ileal mucosal hyperpermeability. The lowest effective concentration for XJB-5-208 and XJB-5-131 was 1 mol/L; ie, both of these compounds were ⬃1000-fold more potent than TEMPOL. Two other compounds carrying the TEMPO payload, XJB-5-125 and XJB-5-197, failed to provide protection against gut barrier dysfunction induced by hemorrhage. XJB5-133 has the same (hemigramicidin-based) mitochondrial targeting moiety as XJB-5-131 but lacks the TEMPO payload. It is noteworthy, therefore, that XJB-5-133 (Fig. 2F) did not afford protection from the development of ileal mucosal hyperpermeability. Ineffective as well were the 2 other hemigramicidin-based compounds that also lacked the TEMPO payload, XJB-5-127 and 194 (Fig. 2D, H). Of the compounds screened, XJB-5-131 appeared to be the most effective,
FIGURE 1. Chemical structures of TEMPOL and the 7 gramicidin S-based compounds screened for the ability to ameliorate hemorrhage-induced ileal mucosal hyperpermeability in rats.
308
© 2007 Lippincott Williams & Wilkins
Annals of Surgery • Volume 245, Number 2, February 2007
Hemigramicidin-TEMPO Conjugate
FIGURE 2. Relative change in ileal mucosal permeability to FD4 in rats subjected to HS. Isolated ileal segments were loaded inside the lumen with the indicated concentrations of TEMPOL or one of 7 different gramicidin S-based compounds. After 2 hours of shock, the gut segments were harvested and mucosal permeability to FD4 measured ex vivo. Data are expressed as a percentage of the change permeability relative to that observed in simultaneously assayed control segments loaded during shock with normal saline solution (Methods). Results are means ⫾ SE (n ⫽ 6 per condition). *P ⬍ 0.05 versus “no treatment” controls.
reducing HS-induced mucosal hyperpermeability to approximately 60% of the control value.
XJB-5-131 Ameliorates Peroxidation of Mitochondrial Phospholipids and Activation of Caspases 3/7 in Gut Mucosa From Rats Subjected to HS To obtain additional information regarding the mechanism(s) responsible for the protective effect of XJB-5-131, we carried out another series of in vivo experiments. Again, isolated segments of intestine were prepared and filled with either vehicle or a 10-mol/L solution of XJB-5-131, the most active of the hemigramicidin-TEMPO conjugates examined in the previous series of experiments. The rats were subjected to 2 hours of HS, and at the end of this period, samples of ileal mucosa from the gut sacs filled with vehicle or XJB-5-131 solution were obtained. Samples of ileal mucosa from completely normal rats were also obtained for purposes of comparison. The samples were assayed for caspase 3/7 activity as well as peroxidation of cardiolipin (CL) and several other phospholipids. When mucosal samples from normal animals were compared with those from vehicle-treated segments from shocked rats, it was apparent that HS was associated with © 2007 Lippincott Williams & Wilkins
significant peroxidation of key phospholipids, including phosphatidylcholine (PC; Fig. 3A), phosphatidylethanolamine (PE; Fig. 3B), phosphatidylserine (PS; Fig. 3C), and CL (Fig. 3D). Treatment with XJB-5-131 significantly ameliorated HS-induced peroxidation of CL, the only 1 of the 4 phospholipids found exclusively in mitochondria. In contrast, treatment with XJB-5-131 had only a small effect on PE peroxidation and no effect on peroxidation of PC and PS. These data indicate that HS is associated with substantial oxidative stress even in the absence of resuscitation (reperfusion). Furthermore, these data support the view XJB-5-131 is an effective ROS scavenger, which localizes predominantly in mitochondria where it is capable of protecting CL from peroxidation. Relative to the activity measured in samples from normal animals, the activity of caspases 3 and 7 was markedly increased in vehicle-treated mucosal samples from hemorrhaged rats (Fig. 4). However, when the ileal segments were filled with XJB-5-131 solution instead of its vehicle, the level of caspase 3 and 7 activity after HS was significantly decreased. Consistent with previously reported findings,29,30 our observations support the view that HS is associated with activation of pro-apoptotic pathways in gut mucosal cells.
309
Macias et al
Annals of Surgery • Volume 245, Number 2, February 2007
FIGURE 3. Effect of intraluminal XJB-5-131 on HS-induced peroxidation of phospholipids in intestinal mucosa. Tissue samples were harvested after 2 hours of shock and assayed for phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), and cardiolipin (CL) as described in Methods. Samples were obtained from intestinal segments of normal rats not subjected to HS (control; n ⫽ 3), intestinal segments filled with vehicle from rats subjected to HS (n ⫽ 5), and intestinal segments filled with 10 mol/L XJB-5-131 from rats subjected to HS (n ⫽ 5). Data are mean ⫾ SE. #P ⬍ 0.05 versus control. *P ⬍ 0.05 versus HS ⫹ vehicle.
XJB-5-131 Ameliorates Oxidant-Induced Epithelial Hyperpermeability In Vitro
FIGURE 4. Effect of intraluminal XJB-5-131 on activation of caspases 3 and 7 in intestinal mucosa. Tissue samples were harvested after 2 hours of shock and assayed for caspase 3/7 activity. Samples were obtained from intestinal segments from normal rats not subjected to HS (control; n ⫽ 3), intestinal segments filled with vehicle from rats subjected to HS (n ⫽ 5), and intestinal segments filled with 10 mol/L XJB5-131 from rats subjected to HS (n ⫽ 5). Data are mean ⫾ SE. #P ⬍ 0.05 versus control. *P ⬍ 0.05 versus HS ⫹ vehicle.
Moreover, our data support the view that this process is significantly ameliorated following (local) treatment with XJB-5-131.
310
Prompted by our observation that XJB-5-131 ameliorated HS-induced CL peroxidation in mucosal cells in vivo, we sought to determine if treatment with this compound could ameliorate menadione-induced epithelial hyperpermeability in vitro. Caco-2BBe monolayers were incubated in the absence or presence of menadione. In some cases, graded concentrations of XJB-5-131 were added as well. As expected, incubation of Caco-2BBe monolayers with menadione for 6 hours caused a marked increase in the apical-basolateral clearance of FD4 (Fig. 5). Treatment with 10 mol/L XJB5-131 provided significant protection against menadioneinduced hyperpermeability. When the concentration of XJB5-131 was increased to 100 mol/L, significant protection was no longer observed, suggesting that this compound has the potential to induce toxic effects at higher concentrations.
Treatment With XJB-5-131 Prolongs Survival of Rats With Lethal HS Since XJB-5-131 had salutary effects on several biochemical and physiologic read-outs in the previous in vivo and in vitro assay systems, we sought to determine whether systemic administration of this compound could prolong the survival of rats subjected to massive blood loss, even in the absence of standard resuscitation with blood and crystalloid © 2007 Lippincott Williams & Wilkins
Annals of Surgery • Volume 245, Number 2, February 2007
Hemigramicidin-TEMPO Conjugate
FIGURE 5. Effect of XJB-5-131 on the permeability of Caco-2BBe human enterocyte-like monolayers subjected to oxidative stress. Monolayers were incubated in the absence or presence of 50 mol/L menadione for 6 hours. In some cases, the apical and basolateral compartments of randomly selected wells were loaded with the indicated concentrations of XJB5-131 or the vehicle solution. The data shown are from a representative experiment that was repeated 3 times with similar results. Data are mean ⫾ SE (n ⫽ 4). #P ⬍ 0.05 versus control. *P ⬍ 0.05 versus menadione.
solution. Sixteen rats were used. Three died during the hemorrhage phase of the protocol (ie, during the first 60 minutes) and were excluded from data analyses. Thirteen survived for at least 60 minutes and received the full dose of either XJB-5-131 solution or vehicle. In both groups, MAP decreased precipitously during the first phase of the hemorrhage protocol and remained nearly constant at 40 mm Hg during the beginning of the second phase. Shortly after treatment was started, MAP increased slightly in both groups (Fig. 6A). Blood glucose, lactate, and hemoglobin concentrations were similar in both groups at baseline and before and immediately after treatment (Table 1). Six of 7 animals in the vehicle-treated (control) group died within 1 hour after the end of the bleeding protocol and all were dead within 125 minutes (Fig. 6B). The rats treated with XJB-5-131 survived significantly longer (P ⬍ 0.01). Three of six survived for longer than 3 hours after completion of the hemorrhage protocol and one rat survived for the whole 6 hours postbleeding observation period.
DISCUSSION Prompted by the recognition that mitochondria are a major intracellular source of ROS under pathologic conditions, there is increasing interest in the notion of developing mitochondria-targeted antioxidants as therapeutic agents.31 Two such compounds, namely, 关2-(3,4-dihydro-6-hydroxy2,5,7,8-tetramethyl-2H-1-benzopyran-2-yl)ethyl兴 triphenylphosphonium bromide (MitoVit E) and 10-(6⬘-ubiquinolyl)decyltriphenylphosphonium bromide (Mitoquinol or MitoQ), have been extensively evaluated in a variety of in vitro models of redox stress. For example, MitoVit E was shown to protect isolated rat liver mitochondria oxidative damage induced by tert-butylhydroperoxide32 and MitoQ was shown to protect Jurkat cells from hydrogen peroxideinduced apoptosis.33 Importantly, Adlam et al recently reported that pretreating rats for 14 days with oral MitoQ resulted in preservation of myocardial function when hearts were studied ex vivo, using a Langendorff constant pressure system, and subjected to transient coronary ischemia and reperfusion.34 Herein, we extended this line of © 2007 Lippincott Williams & Wilkins
investigation by showing for the first time that acute administration of a single dose of a mitochondria-targeted ROS scavenger, XJB-5-131, can have dramatic salutary physiologic and biochemical effects in an animal model of critical illness. CL is an anionic phospholipid found exclusively in the inner mitochondrial membrane of eukaryotic cells.35 Under normal conditions, the pro-apoptotic protein, cytochrome c, is anchored to the mitochondrial inner membrane by its specific and stoichiometric association with CL.36 The acyl chains of CL are unsaturated and, therefore, are susceptible to peroxidation by reactive oxygen species (ROS), such as peroxynitrite (ONOO-), which is formed when O2·⫺ reacts with nitric oxide. When ROS are generated within mitochondria in excessive quantities, cytochrome c bound to CL can function as an oxidase and induce extensive peroxidation of CL in the inner mitochondrial membrane.26,37 Peroxidation of CL has 2 important consequences. First, peroxidized CL fails to bind cytochrome c tightly,38 leading to release of this protein into the intermembrane space. Second, peroxidation of CL is important for opening of the mitochondrial permeability transition pore.39,40 Opening of the mitochondrial permeability transition pore promotes mitochondrial swelling and release into the cytosol of cytochrome c. Thus, peroxidation of CL promotes the release of cytochrome c and, on this basis, apoptosis.35 The findings presented here support the view that HS promotes the peroxidation of CL and that treatment with an ROS and electron scavenger, such as XJB-5-131, can ameliorate HS-induced peroxidation of CL. Since peroxidation of CL is known to be pro-apoptotic, one would predict that treatment with XJB-5-131 could protect against HSinduced cellular apoptosis. Indeed, using activation of caspases 3/7 as a marker for apoptosis, we were able to show that HS increased activation of these key enzymes in ileal mucosal cells, but treatment with XJB-5-131 significantly ameliorated this effect. The findings presented here suggest that treatment with XJB-5-131 might be able to prolong the period of time that patients can survive after losing large quantities
311
Annals of Surgery • Volume 245, Number 2, February 2007
Macias et al
FIGURE 6. Effects of intravenous treatment with XJB-5-131 on MAP (A) and survival (B) of rats subjected to volumecontrolled HS. Rats were treated with 2.8 mL/kg of vehicle or the same volume of XJB-5-131 solution during the final 20 minutes of the bleeding protocol. The total dose of XJB-5-131 infused was 2 mol/kg. The difference in survival time between the 2 groups was statistically significant (P ⬍ 0.01).
TABLE 1. Blood Glucose, Lactate, and Hemoglobin Concentrations in Hemorrhaged Rats at Baseline, at the End of the First Phase of the Hemorrhage Protocol, and Then at End of the Second Phase of the Hemorrhage Protocol 关ie, immediately after treatment with either vehicle (n ⫽ 7) or XJB-5-131 (n ⫽ 6)兴 Parameter
Category
Baseline
End of First Phase of Hemorrhage
Blood glucose concentration (mg/dL)
Vehicle XJB-131 Vehicle XJB-131 Vehicle XJB-131
143 ⫾ 5 134 ⫾ 4 1.8 ⫾ 0.4 1.8 ⫾ 0.2 12.7 ⫾ 0.5 12.7 ⫾ 0.3
255 ⫾ 30 228 ⫾ 24 6.6 ⫾ 0.8 5.7 ⫾ 0.8 11.1 ⫾ 0.3 10.7 ⫾ 0.3
Blood lactate concentration (mEq/L) Blood Hb concentration (g/dL)
End of Second Phase of Hemorrhage 219 ⫾ 26 201 ⫾ 38 5.9 ⫾ 1.3 5.6⫾1.2 9.4 ⫾ 0.2 9.4 ⫾ 0.3
Data are mean ⫾ SE. None of the between-group differences was statistically significant.
312
© 2007 Lippincott Williams & Wilkins
Annals of Surgery • Volume 245, Number 2, February 2007
of blood due to traumatic injuries or other catastrophes (eg, rupture of an abdominal aortic aneurysm). By extending the treatment window before irreversible shock develops, treatment in the field with XJB-5-131 might “buy” enough time to allow transport of more badly injured patients to locations where definitive care, including control of bleeding and resuscitation with blood products and asanguinous fluids, can be provided. Our findings using a rodent model of HS also open up the possibility that drugs like XJB-5131 might be beneficial in other conditions associated with marked tissue hypoperfusion, such as stroke and myocardial infarction. The findings presented here also support the general concept that mitochondrial targeting of ROS scavengers is a reasonable therapeutic strategy for the management of HS. Although previous studies have shown that treatment with TEMPOL is beneficial in rodent models of HS,27,28 a relatively large dose (175 mol/kg bolus ⫹ 175 mol/kg per h) of the compound was required. In contrast, treatment with a dose of XJB-5-131 (about 2 mol/kg) that was more than 80-fold smaller was clearly beneficial. The greater potency of XJB-5-131 as compared with TEMPOL presumably reflects the tendency of the former compound to localize in mitochondrial membranes. We have previously verified the localization of hemigramicidin-TEMPO conjugates using electron paramagnetic resonance (EPR) spectroscopy. Because 4-NH2-TEMPO is a stable free radical, the presence of the compound can be assessed in biologic samples because of its characteristic triplet EPR signal. Using this approach, we showed that 2 hemigramicidin-TEMPO conjugates are concentrated in the mitochondria of cultured mouse embryonic cells following incubation of the cells with 10 mol/L solutions of the compounds.14 In contrast, 4-NH2-TEMPO did not effectively partition into either cells or mitochondria.14 Previously, we reported that 2 hemigramicidin-TEMPO conjugates are capable of protecting mouse embryonic cells from caspase activation and apoptosis induced by incubation with actinomycin D.14 Herein, we extended this line of in vitro investigation by showing that XJB-5-131 preserves barrier function when Caco-2 human enterocyte-like monolayers are incubated with the redox-cycling ROS generator, menadione. The permeability assay using Caco-2 cells growing in bicameral diffusion chambers is straightforward and uses a fluorometric read-out. Accordingly, this assay system might prove useful for screening relatively large libraries of mitochondria-targeted antioxidants for cytoprotective activity. Our study has a number of important limitations. For example, at this early stage in our investigation of XJB-5-131 and related compounds, we know nothing about the half-life, distribution, metabolism, or elimination of these novel therapeutic agents. Accordingly, the dose of XJB-5-131 that we used for our in vivo experiments (2 mol/kg) was simply a best guess based on the results of our preliminary experiments using the gut sac model system. Higher or lower doses, repetitive dosing, or use of a continuous infusion might be more effective. © 2007 Lippincott Williams & Wilkins
Hemigramicidin-TEMPO Conjugate
We have not carried out any studies to assess the toxicity XJB-5-131. However, we know that both TEMPOlike compounds and gramicidin S are capable of causing a number of deleterious effects, ranging from hypotension41 to hemolysis.42 Since the molecular structure of XJB-5-131 makes it both a TEMPO-like compound and a derivative of gramicidin S, the potential for toxicity is quite real. By the same token, however, no deleterious effects were detected in our in vivo studies of the compound in rats subjected to HS. Another important limitation of our study is that we examined the effects of XJB-5-131 only in rats that were subjected to shock but never fully resuscitated with blood or asanguinous fluids. Although ischemia during the shock phase can injure tissues, it is well established that reperfusion also contributes to cellular damage. It remains to be determined whether early treatment with XJB-5-131 to prolong survival during the shock phase followed by standard resuscitation with blood and crystalloid solution will lead to improved long-term survival. These studies will be carried out in the near future. Finally, we performed all of our in vivo studies using rats that were anesthetized with ketamine and sodium pentobarbital. We recognize that these anesthetic agents have numerous pharmacologic effects and that treatment with XJB-5-131 might be more or less beneficial in unanesthetized animals or humans.
CONCLUSION XJB-5-131 is a novel ROS and electron scavenger that is concentrated in mitochondria. Herein, we showed that treatment with this compound in the absence of conventional resuscitation with blood or asanguinous fluid can prolong survival of rats subjected to lethal HS. If future studies verify that treatment with this compound in combination with conventional resuscitation can provide long-term protection against HS-induced lethality, then additional pharmacologic and toxicologic studies will be warranted to determine whether this agent (or other related compounds) should be further developed for clinical testing. REFERENCES 1. Anderson RN, Smith BL. Deaths: leading causes for 2002. National Vital Statistics Reports. 2005;53:1–90. 2. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma. 1995;38:185–193. 3. Acosta JA, Yang JC, Winchell RJ, et al. Lethal injuries and time to death in a level I trauma center. J Am Coll Surg. 1998;186:528 –533. 4. Owens TM, Watson WC, Prough DS, et al. Limiting initial resuscitation of uncontrolled hemorrhage reduces internal bleeding and subsequent volume requirements. J Trauma. 1995;39:200 –207. 5. Bickell WH, Wall MJ Jr, Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med. 1994;331:1105–1109. 6. Shah KJ, Chiu WC, Scalea TM, et al. Detrimental effects of rapid fluid resuscitation on hepatocellular function and survival after hemorrhagic shock. Shock. 2002;18:242–247. 7. Kulisz A, Chen N, Chandel NS, et al. Mitochondrial ROS initiate phosphorylation of p38 MAP kinase during hypoxia in cardiomyocytes. Am J Physiol Lung Cell Mol Physiol. 2002;282:L1324 –L1329. 8. Guzy RD, Hoyos B, Robin E, et al. Mitochondrial complex III is required for hypoxia-induced ROS production and cellular oxygen sensing. Cell Metab. 2005;1:401– 408. 9. Park Y, Kehrer JP. Oxidative changes in hypoxic-reoxygenated rabbit
313
Macias et al
10. 11. 12. 13. 14. 15. 16. 17. 18.
19. 20. 21.
22.
23. 24.
25. 26.
heart: a consequence of hypoxia rather than reoxygenation. Free Radic Res Comm. 1991;14:179 –185. Krishna MC, Russo A, Mitchell JB, et al. Do nitroxide antioxidants act as scavengers of O2·⫺. or as SOD mimics? J Biol Chem. 1996;271: 26026 –26031. Samuni A, Mitchell JB, DeGraff W, et al. Nitroxide SOD-mimics: modes of action. Free Radic Res Commun. 1991;12:187–194. Samuni A, Krishna CM, Mitchell JB, et al. Superoxide reaction with nitroxides. Free Radic Res Commun. 1990;9:241–249. Scholtz KF, Solovjena NA, Kotelnikova AV, et al. Effect of gramicidin S and its derivatives on the mitochondrial membrane. FEBS Lett. 1975;58:141–144. Wipf P, Xiao J, Jiang J, et al. Mitochondrial targeting of selective electron scavengers: synthesis and biological analysis of hemigramicidin-TEMPO conjugates. J Am Chem Soc. 2005;127:12460 –12461. Wattanasirichaigoon S, Menconi MJ, Fink MP. Lisofylline ameliorates intestinal and hepatic injury induced by hemorrhage and resuscitation in rats. Crit Care Med. 2000;28:1540 –1549. Wattanasirichaigoon S, Menconi MJ, Delude RL, et al. Effect of mesenteric ischemia and reperfusion or hemorrhagic shock on intestinal mucosal permeability and ATP content in rats. Shock. 1999;12:127–133. Yang R, Gallo DJ, Baust JJ, et al. Ethyl pyruvate modulates inflammatory gene expression in mice subjected to hemorrhagic shock. Am J Physiol Gastrointest Liver Physiol. 2002;283:G212–G22. Sappington PL, Han X, Yang R, et al. Ethyl pyruvate ameliorates intestinal epithelial barrier dysfunction in endotoxemic mice and immunostimulated Caco-2 enterocytic monolayers. J Pharmacol Exp Ther. 2003;304:464 – 476. Sappington PL, Yang R, Yang H, et al. HMGB1 B box increases the permeability of Caco-2 enterocytic monolayers and impairs intestinal barrier function in mice. Gastroenterology. 2002;123:790 – 802. Baker RD, Baker SS, LaRosa K. Polarized Caco-2 cells: effect of reactive oxygen metabolites on enterocyte barrier function. Digestive Dis Sci. 1995;40:510 –518. Banan A, Fields JZ, Farhadi A, et al. Activation of delta-isoform of protein kinase C is required for oxidant-induced disruption of both the microtubule cytoskeleton and permeability barrier of intestinal epithelia. J Pharmacol Exp Ther. 2002;303:17–28. Han X, Fink MP, Delude RL. Proinflammatory cytokines cause · NO⫺ dependent and independent changes in expression and localization of tight junction proteins in intestinal epithelial cells. Shock. 2003;19: 229 –237. Folch J, Lees M, Sloan-Stanley GH. A simple method for isolation and purification of total lipids from animal tissue. J Biol Chem. 1957;226: 497–509. Kagan VE, Gleiss B, Tyurina YY, et al. A role for oxidative stress in apoptosis: oxidation and externalization of phosphatidylserine is required for macrophage clearance of cell undergoing Fas-mediated apoptosis. J Immunol. 2002;169:487– 489. Bottcher CJF, Van Gent CM, Pries C. A rapid and sensitive sub-micro phosphorus determination. Anal Chim Acta. 1961;24:203–204. Kagan VE, Tyurin VA, Jiang J, et al. Cytochrome c acts as a cardiolipin
314
Annals of Surgery • Volume 245, Number 2, February 2007
27.
28.
29.
30.
31.
32. 33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
oxygenase required for release of proapoptotic factors. Nature Chem Biol. 2005;1:223–232. Paxian M, Bauer I, Rensing H, et al. Recovery of hepatocellular ATP and ‘pericentral apoptosis’ after hemorrhage and resuscitation. FASEB J. 2003;17:993–1002. Mota-Filipe H, McDonald MC, Cuzzocrea S, et al. A membranepermeable radical scavenger reduces the organ injury in hemorrhagic shock. Shock. 1999;12:255–261. Rollwagen FM, Yu ZY, Li YY, et al. IL-6 rescues enterocytes from hemorrhage induced apoptosis in vivo and in vitro by a bcl-2 mediated mechanism. Clin Immunol Immunopathol. 1998;89:205–213. Chang JX, Chen S, Ma LP, et al. Functional and morphological changes of the gut barrier during the restitution process after hemorrhagic shock. World J Gastroenterol. 2005;11:5485–5491. Sheu SS, Nauduri D, Anders MW. Targeting antioxidants to mitochondria: a new therapeutic direction. Biochim Biophys Acta. 2006;1762: 256 –265. Smith RA, Porteous CM, Coulter CV, et al. Selective targeting of an antioxidant to mitochondria. Eur J Biochem. 1999;263:709 –716. Kelso GF, Porteous CM, Coulter CV, et al. Selective targeting of a redox-active ubiquinone to mitochondria within cells: antioxidant and antiapoptotic properties. J Biol Chem. 2001;276:4588 – 4596. Adlam VJ, Harrison JC, Porteous CM, et al. Targeting an antioxidant to mitochondria decreases cardiac ischemia-reperfusion injury. FASEB J. 2005;19:1088 –1095. Iverson SL, Orrenius S. The cardiolipin-cytochrome c interaction and the mitochondrial regulation of apoptosis. Arch Biochem Biophys. 2003; 423:37– 46. Tuominen EKJ, Wallace CJA, Kinnunen PKJ. Phospholipid-cytochrome c interaction: evidence for the extended lipid anchorage. J Biol Chem. 2002;277:8822– 8826. Kagan VE, Borisenko GG, Tyurina YY, et al. Oxidative lipidomics of apoptosis: redox catalytic interactions of cytochrome c with cardiolipin and phosphatidylserine. Free Radic Biol Med. 2004;37:1963–1985. Shidoji Y, Hayashi K, Komura S, et al. Loss of molecular interaction between cytochrome c and cardiolipin due to lipid peroxidation. Biochem Biophys Res Commun. 1999;264:343–347. Dolder M, Wendt S, Walliman T-F. Mitochondrial creatine kinase in contact sites: interaction with porin and adenine nucleotide translocase, role in permeability transition and sensitivity to oxidative damage. Biol Signals Recept. 2001;10:93–111. Imai H, Koumura T, Nakajima R, et al. Protection from inactivation of the adenine nucleotide translocator during hypoglycaemia-induced apoptosis by mitochondrial phospholipid hydroperoxide glutathione peroxidase. Biochem J. 2003;371:799 – 809. Hahn SM, Sullivan FJ, DeLuca AM, et al. Hemodynamic effect of the nitroxide superoxide dismutase mimics. Free Radic Biol Med. 1999;27: 529 –535. Kondejewski LH, Farmer SW, Wishart DS, et al. Gramicidin S is active against both gram-positive and gram-negative bacteria. Int J Pept Protein Res. 1996;47:460 – 466.
© 2007 Lippincott Williams & Wilkins
