ORIGINAL ARTICLE
Remote Ischemic Conditioning Prevents Lung and Liver Injury After Hemorrhagic Shock/Resuscitation Potential Role of a Humoral Plasma Factor Chung Ho Leung, BSc,∗ †‡ Christopher A. Caldarone, MD,†‡ Feng Wang, MSc,∗ Seetha Venkateswaran, BSc,∗ Menachem Ailenberg, PhD,∗ Brian Vadasz, BSc,∗ ‡ Xiao-Yan Wen, PhD,∗ ‡ and Ori D. Rotstein, MD∗ ‡ Objective: To evaluate the efficacy of remote ischemic conditioning (RIC) on organ protection after hemorrhagic shock/resuscitation (S/R) in a murine model. Background: Ischemia/reperfusion resulting from S/R contributes to multiple organ dysfunction in trauma patients. We hypothesized that RIC before shock (remote ischemic preconditioning), during shock (remote ischemic “PER”conditioning), or during resuscitation (remote ischemic “POST”conditioning) could confer organ protection. We also tested the effect of ischemic conditioned plasma on neutrophil migration in vivo using transgenic zebrafish models. Methods: C57Bl/6 mice were subjected to S/R with or without hindlimb RIC. Serum levels of alanine aminotransferase and tumor necrosis factor-alpha, and liver tumor necrosis factor-alpha and interleukin 1β mRNA were evaluated. In some experiments, lung protein leakage, cytokines, and myeloperoxidase activity were investigated. Plasma from mice subjected to RIC was microinjected into zebrafish, and neutrophil migration was assessed after tailfin transection or copper sulfate treatment. Results: In mice subjected to S/R, remote ischemic preconditioning, remote ischemic “PER”conditioning, and remote ischemic “POST”conditioning each significantly reduced serum alanine aminotransferase and liver mRNA expression of tumor necrosis factor-alpha and interleukin 1β and improved liver histology compared with control S/R mice. Lung injury and inflammation were also significantly reduced in mice treated with remote ischemic preconditioning. Zebrafish injected with plasma or dialyzed plasma (fraction >14 kDa) from ischemic conditioned mice had reduced neutrophil migration toward sites of injury compared with zebrafish injected with control plasma. Conclusions: RIC protects against S/R-induced organ injury, in part, through a humoral factor(s), which alters neutrophil function. The beneficial effects of RIC, performed during the S/R phase of care, suggest a role for its application early in the posttrauma period. Keywords: hemorrhagic shock, ischemia/reperfusion, remote ischemic conditioning, resuscitation, trauma (Ann Surg 2015;261:1215–1225)
H
emorrhagic shock and resuscitation (S/R) after traumatic injury contributes to the development of the systemic inflammatory response and multiorgan dysfunction in this patient population.1 Several mechanisms have been invoked to explain this response to From the ∗ Department of Surgery, the Keenan Research Centre for Biomedical Science and the Zebrafish Centre for Advanced Drug Discovery, St. Michael’s Hospital, Toronto, Ontario, Canada; †Division of Cardiovascular Surgery, Hospital for Sick Children, Toronto, Ontario, Canada; and ‡Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada. Disclosure: Supported by the Canadian Institutes of Health Research grant #37779. The authors declare no conflicts of interest. Reprints: Ori D. Rotstein, MD, St. Michael’s Hospital, 30 Bond St 16CC-044, Toronto, ON M5B 1W8, Canada. E-mail: [email protected]. C 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright ISSN: 0003-4932/14/26106-1215 DOI: 10.1097/SLA.0000000000000877
Annals of Surgery r Volume 261, Number 6, June 2015
S/R. One unifying theme is that S/R represents a form of global ischemia/reperfusion, which initiates a broad range of cellular signaling events, which directly or indirectly promote cellular injury and inflammation. For example, oxidative stress derived from ischemia/reperfusion not only contributes directly to local organ injury through the generation of injurious reactive oxygen species but also participates in cell signaling events leading to the activation of inflammatory molecules.2,3 In addition, S/R primes immune cells to exhibit an exaggerated phenotype, leading to subsequent lung injury in a second hit.4 Strategies directed at lessening ischemia/reperfusion injury caused by S/R have potential to improve outcome in this patient population. One potent intervention that harnesses the cell’s innate protective mechanism against ischemia/reperfusion injury is known as remote ischemic preconditioning (RIPC, reviewed by Szij´art´o et al5 ). In experimental RIPC in animals, cycles of transient ischemia/reperfusion applied to the limb have been shown to reduce subsequent ischemia/reperfusion injury in multiple distant organs including the heart, liver, lungs, and brain.5 In humans, RIPC also seems to exert beneficial effects. Its effects have been most thoroughly studied in the context of diminishing myocardial injury after surgical or radiological interventions.6 The concept of ischemic conditioning was initially described in reference to a conditioning stimulus applied before ischemia/reperfusion of the target organ (ie, “preconditioning”).7 More recently, the concept of ischemic conditioning has been extended to provide the conditioning stimulus during ischemia of the target organ (“perconditioning”) and after ischemia of the target organ (“postconditioning”). The extension of effective organ conditioning to include these periods broadens the potential clinical use of this protective intervention. Like preconditioning, both per- and postconditioning have received the greatest attention in the cardiac domain, although studies have demonstrated the efficacy of these approaches to ischemia/reperfusion injury in other organs.5,8 In the context of S/R, preconditioning may have potential relevance to the military setting, where combatants at risk of sustaining later injury may submit to RIPC before engaging in combat. In this regard, a single study recently reported by Jan and colleagues9 demonstrated the ability of RIPC to mitigate shock-induced lung injury in a rat model. However, demonstration of effective per- and postconditioning would further extend its use in the trauma setting. Specifically, should these latter strategies prove efficacious, this would open the window for medical personnel to use remote ischemic conditioning during the shock or resuscitation phases of trauma care, with a view to lessening the consequences of S/R. The present studies evaluate the ability of remote ischemic conditioning to reduce liver and lung injury in a murine model of hemorrhagic shock and to investigate potential mechanisms for protection. We show that preconditioning, as well as perconditioning and postconditioning, are each able to mitigate organ injury and lessen inflammation after S/R. Furthermore, we investigated a potential mechanism underlying the beneficial effect of remote ischemic www.annalsofsurgery.com | 1215
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
conditioning. Using the zebrafish to study wound-induced neutrophil migration in vivo,10 we demonstrate that a humoral plasma factor(s) generated by the ischemic conditioning protocol in mice inhibits neutrophil migration to sites of injury.
(A)
Resuscitaon (2-volume saline of shed blood)
Hemorrhagic Shock (22.5 μl blood/g body weight)
Control S/R
METHODS
60’
Animal Models of Hemorrhagic S/R In preliminary studies, we examined the effect of different periods of hemorrhagic shock on the development of liver injury in the reperfusion phase. Consistent with the literature,11 the duration of hypovolemic shock was an important determinant of the magnitude of liver injury. We observed that 1 hour of shock induced mild liver injury whereas a 2-hour hypotensive period produced much more severe injury with alanine aminotransferase (ALT) levels that were approximately 10-fold higher than those with 1 hour of hypotension (data not shown). This afforded us the opportunity to study the effect of ischemic conditioning in models of both mild and severe injuries.
Remote Ischemic Conditioning Protocol in the S/R Model Inducing Mild Liver Injury Adult male C57Bl/6 mice (Charles River Laboratories, St Constant, Quebec), 9 to 11 weeks old, weighing 22 to 28 g, were anesthetized with isoflurane and the right carotid artery was cannulated by a PE-10 microtube for hemorrhagic S/R. Before the initiation of hemorrhagic shock, a cycle of left femoral artery occlusion for 10 minutes, followed by reperfusion for 10 minutes, was performed as a means of inducing RIPC. Control S/R animals had undergone groin dissection and isolation of the femoral artery but without artery occlusion. Hemorrhagic shock was accomplished by a 1-time blood withdrawal (22.5 μL of blood/g body weight) over 15 minutes. The amount of blood withdrawn lowered the mean arterial blood pressure to 30 mm Hg as monitored by a transducer via the carotid cannula. After a hypotensive period of 60 minutes, animals were resuscitated with 0.9% normal saline equivalent to twice the volume of blood withdrawn over 15 minutes. At 1 and 2 hours after resuscitation, liver and lung tissues and blood samples were collected for analysis (Fig. 1A).
Remote Ischemic Conditioning Protocol in the S/R Model Inducing Severe Liver Injury Mice were subjected to hemorrhagic shock at 30 mm Hg of mean arterial blood pressure for 2 hours by blood withdrawal from the left femoral artery. Mean arterial blood pressure was continuously monitored through the left carotid artery over this period. Mice were then resuscitated with shed blood and 2 volumes of lactated Ringer’s solution over 15 minutes. At 2 hours after resuscitation, liver tissue and blood samples were collected for analysis. Using this model, the effect of remote ischemic conditioning on liver injury was investigated. We modified these protocols to more closely mimic that reported both in experimental animals and in humans.8,12 Specifically, RIPC was accomplished by performing 4 cycles of alternating 5-minute ischemia, followed by 5-minute reperfusion of the right lower extremity, using an externally applied tourniquet before shock. Control S/R animals were subjected to anesthesia but without hindlimb ischemia. Given the fact that, in the trauma setting, it would be unlikely to be able to perform preconditioning, we also examined the effect of remote ischemic “PER”conditioning (RIPerC) as well as combined RIPerC and “POST”conditioning (RIPostC) on liver injury. In RIPerC, the 4 cycles are performed during the final 40 minutes of the shock period, whereas in the RIPerC + RIPostC, the first of the 4 cycles is performed during the last 10 minutes of shock and the final 3 cycles are carried into the resuscitation period (Fig. 1B). 1216 | www.annalsofsurgery.com
15’
120’
15’
120’
Remote Ischemic Precondioning (RIPC) 10
60’
10
Hemorrhagic Shock (30 mm HG MAP)
(B)
Resuscitaon (Shed Blood + 2-Volume Ringer’s Lactate)
Control S/R
120’
15’
120’
15’
120’
15’
120’
Remote Ischemic Precondioning (RIPC)
120’
55
Remote Ischemic Percondioning (RIPerC)
120’
55
RIPerC + Remote Ischemic Postcondioning (RIPerC+RIPostC)
120’
15’
120’
55
FIGURE 1. A, Schematic of hemorrhagic S/R protocol of mild liver injury. Remote ischemic preconditioning was performed by 10-minute femoral artery occlusion, followed by 10-minute reperfusion. B, Schematic of hemorrhagic S/R protocol of severe liver injury. Remote ischemic conditioning was performed by 4 cycles of 5-minute ischemia and 5-minute reperfusion by a tourniquet.
Serum ALT ALT levels as a marker of hepatocellular injury were measured in blood samples taken at 1 and 2 hours after resuscitation by the Diagnostic Laboratory at St. Michael’s Hospital.
Liver Histology Liver tissue obtained at 2 hours after resuscitation was fixed by immersion in formaldehyde, embedded by paraffin wax, and cut in 5-μm slices. Sections were stained with hematoxylin and eosin and evaluated by light microscopy in a blinded manner by an independent person using the following scoring criteria13 : 0, no hepatocellular damage; 1, mild injury characterized by cytoplasmic vacuolization and focal nuclear pyknosis; 2, moderate injury with dilated sinusoids, cytosolic vacuolization, and blurring of intercellular borders; 3, moderate-to-severe injury with coagulative necrosis, abundant sinusoidal dilation, red blood cell extravasation into hepatic chords, and hypereosinophilia and margination of neutrophils; 4, severe necrosis with loss of hepatic architecture, disintegration of hepatic cords, hemorrhage, and neutrophil infiltration. C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Limb Conditioning Prevents Organ Injury
Real-time Quantitative Reverse Transcription Polymerase Chain Reaction of Inflammatory Gene mRNA Expression
GCCAAGTATGATGACA and TGAAGTCGCAGGAGACAACCT, respectively.
TNF-α Enzyme-linked Immunosorbent Assay
Total RNA from the liver and lung were extracted using RNeasy Mini Kit (Qiagen, Valencia, CA). One microgram of RNA was then subjected to DNAse treatment and reverse transcription (Bio-rad, Hercules, CA). The relative abundance of mRNA for tumor necrosis factor alpha (TNF-α), interleukin 1β (IL-1β), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, endogenous control) in the liver and lung tissues were quantified using SYBR green-based quantitative polymerase chain reaction (PCR) (Applied Biosystems, Foster City, CA). Each reaction was performed in triplicate. Realtime PCR conditions were as follows: initial denaturation at 95◦ C for 10 minutes, followed by 40 cycles of 95◦ C for 15 seconds, and 60◦ C for 60 seconds. Purity of the PCR reaction was verified by the meltcurve analysis at the completion of the PCR. Relative abundance of gene expression was calculated by the 2−CT method, using GAPDH as the endogenous control and one of the sham-operated samples as reference. All primers were purchased from Sigma-Aldrich, Oakville, Ontario, Canada. The sequences of forward and reverse primers for TNF-α are TATGGCTCAGGGTCCAACTC and CTCCCTTTGCAGAACTCAGG, respectively. The sequences of forward and reverse primers for IL-1β are GCCCATCCTCTGTGACTCAT and AGGCCACAGGTATTTTGTCG, respectively. The sequences of forward and reverse primers for GAPDH are AGAAACCT-
Serum levels of TNF-α were quantitated in blood samples taken at 1 and 2 hours into resuscitation using commercially available enzyme-linked immunosorbent assay kits according to the manufacturer’s instructions (R&D Systems, Minneapolis, MN).
TNF-α Western Blots Liver and lung tissue protein homogenates were separated via 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Bio-Rad Laboratories, Inc., Hercules, CA). The membranes were then blocked with 5% nonfat milk and probed with TNF-α rat antibody (BioLegend, San Diego, CA) and GAPDH rabbit antibody (Cell Signaling, Danvers, MA) for loading control, followed by appropriate secondary antibodies conjugated with horseradish peroxidase. The blots were exposed to film after enhanced chemiluminescence (Amersham Biosciences, Piscataway, NJ). Densitometry of the blots was analyzed by Quantity One (Bio-Rad Laboratories, Inc., Hercules, CA).
Bronchoalveolar Lavage At the end of resuscitation, tracheostomy was performed and the lungs were lavaged 3 times with cold phosphate-buffered saline in
(A)
(B)
80
Serum ALT (U/L)
70
P < 0.01
Liver TNF-α mRNA (Fold over Sham)
P < 0.05
90
P < 0.01
60 50 40 30 20 10
8
P < 0.05
7 6 5 4 3 2 1 0
0 Sham Operated Control S/R 1 h
RIPC S/R 1 h
Control S/R 2 h
P < 0.01 (C)
Sham operated
RIPC S/R 2 h
Control S/R 2 h
RIPC S/R 2 h
P < 0.05
Liver IL-1β mRNA (Fold over Sham)
30 25 20 15 10 5 0 Sham operated
Control S/R 2 h
RIPC S/R 2 h
FIGURE 2. A, In the mild liver injury model, S/R resulted in liver injury as indicated by increased serum ALT levels at 1 and 2 hours after resuscitation in control S/R mice. Serum ALT was significantly reduced in mice with RIPC. n = 5–8 per group. B, mRNA expression of TNF-α evaluated by quantitative real-time-PCR in the liver after S/R. n = 4–5 per group. C, IL-1β mRNA expression in the liver after S/R. n = 4–5 per group. C 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 1217
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
(B)
3
Liver Histology Injury Score
(A)
2.5
P < 0.01
2 1.5 1 0.5 0 Control S/R 2 h
RIPC S/R 2 h
FIGURE 3. A, Liver histology imaged at 40× magnification by hematoxylin and eosin staining after S/R. Compared with shamoperated mice, S/R significantly increased overall hepatic ballooning degeneration and loss of sinusoidal architecture. However, these were reduced in mice with RIPC. B, Liver injury histology score is significantly lower in mice treated with RIPC. n = 5 per group.
800-μL aliquots. Total protein content in the bronchoalveolar lavage fluid (BAL) fluid as an indicator of lung injury was assayed using a protein assay kit (Bio-Rad Laboratories, Inc., Hercules, CA).
Lung tissue was homogenized and sonicated in 0.5% hexadecyltrimethylammonium bromide (Sigma, Oakville, Ontario, Canada) in 50 mM potassium phosphate buffer (pH 6.0) and assayed with 0.167 mg/mL of o-dianisidine (Sigma, Oakville, Ontario, Canada) and 0.005% hydrogen peroxide. The change in absorbance was measured at 460 nm (ε = 1.3 × 104 M−1 ·cm−1 ). Myeloperoxidase (MPO) activity is expressed as units of MPO per milligram of lung protein, whereby 1 unit of MPO is defined as the amount of enzyme degrading 1 nmol hydrogen peroxide per minute.
P < 0.01
P < 0.01 P < 0.05
600 Serum TNF-α (pg/mL)
Myeloperoxidase Enzyme Activity
700
500 400 300 200 100 0 Sham Operated Control S/R 1 h
RIPC S/R 1 h
Control S/R 2 h
RIPC S/R 2 h
FIGURE 4. Serum levels of TNF-α were significantly increased after S/R and this was significantly prevented in mice with RIPC. n = 3–5 per group.
Zebrafish Models of In Vivo Neutrophil Infiltration We used a transgenic zebrafish [Tg(mpx::GFP)] model with specific green fluorescent protein (GFP)-tagged neutrophils (GFP controlled by the MPO gene promoter) to monitor, in real time, the migration of neutrophils to the wounded area in response to injury induced by tailfin transection.14 We also used a chemical injury model in a separate transgenic zebrafish line (cldnB::GFP/lysC::DsRED2) by copper sulfate treatment, which induces an inflammatory process characterized by migration of neutrophils, labeled with red fluorescent protein (DsRED), to the copper sulfate-injured neuromast cells (labeled with GFP).15 1218 | www.annalsofsurgery.com
Preparation of Remote Ischemic Conditioning Plasma for Zebrafish Microinjection For these studies, we obtained blood from the mice after 4 cycles of hindlimb ischemic conditioning (HIC) to study the effect of plasma on neutrophil function using zebrafish models of neutrophil migration. Blood was collected at 15 minutes after the last cycle of conditioning into sodium citrate tubes for plasma collection. To determine a molecular weight cutoff for humoral factor(s) responsible for conditioning, control and HIC plasma were dialyzed in a C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Limb Conditioning Prevents Organ Injury
(A)
(B)
P < 0.01
P < 0.01 0.8 0.7
P < 0.01
P < 0.01
P < 0.01
P < 0.05
90
0.6
MPO U/mg lung proen
BAL Protein (mg/mL)
80 0.5 0.4 0.3 0.2 0.1
70 60 50 40 30 20 10
0 Sham Operated Control S/R 1 h
RIPC S/R 1 h
Control S/R 2 h
RIPC S/R 2 h
0 Sham Operated
P < 0.01
(C)
(D)
P < 0.01
P < 0.01
20 Lung IL-1β mRNA (Fold over Sham)
Lung TNF-α mRNA (Fold over Sham)
4 3.5 3 2.5 2 1.5 1 0.5
Control S/R 2 h
RIPC S/R 2 h
P < 0.01
18 16 14 12 10 8 6 4 2 0
0
Sham operated
Sham operated
Control S/R 2 h
Control S/R 2 h
RIPC S/R 2 h
RIPC S/R 2 h
(E) Sham Operated
Control S/R 2 h
RIPC S/R 2h
TNF-α
GAPDH
3
P < 0.05
P < 0.01
Lung TNF-α/GAPDH
2.5 2 1.5 1 0.5 0 Sham Operated
Control S/R 2 h
RIPC S/R 2 h
FIGURE 5. A, Lung injury as indicated by protein leakage in the BAL is significantly increased in control mice after S/R. RIPC prevented this increase. n = 3–5 per group. B, MPO enzymatic activity in lung tissue as an indicator of neutrophil infiltration is significantly increased at 2 hours after resuscitation in control mice. RIPC prevented this increase. n = 4–6 per group. C, TNF-α mRNA expression in the lung after S/R. D, IL-1β mRNA expression in the lung after S/R. n = 5 per group. E, Western blot analysis of TNF-α protein expression in lung tissue at 2 hours after resuscitation. n = 3 per group.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 1219
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
14-kDa cutoff dialysis membrane (Spectrum, Rancho Dominguez, CA) overnight.16
(A) 900 800
cldnB::GFP and lysC::DsRED2 Zebrafish: Chemical Injury by Copper Sulfate cldnB::GFP zebrafish expressing GFP on neuromast cells were crossed with lysC::DsRED2 zebrafish expressing DsRED on neutrophils to produce larvae expressing both fluorescent proteins. Fourday postfertilization larvae were injected with control or HIC dialysate (>14 kDa) and were incubated for 24 hours before treatment with copper sulfate (1 μM) for 1.5 hours (peak neutrophil count). Neutrophil migration localized to the lateral line neuromast cell areas was quantified.
Serum ALT (U/L)
Three days postfertilization Tg(mpx::GFP) larvae were anesthetized by immersion in clove oil. To evaluate the effect of HIC on neutrophils, larvae were injected with vehicle (saline), control, or HIC plasma (1:20 dilution in saline), or the dialyzed fractions (>14 kDa and 14 kDa) significantly reduced neutrophil counts in the wounded area compared with the control dialysate (>14 kDa) (P < 0.004). In contrast, neutrophil counts after injection with HIC dialysate (14 kDa
HIC Dialysate >14 kDa
FIGURE 9. A, Fluorescent image of an untreated cldnB::GFP/lysC::DsRED2 zebrafish. Open arrow represents a lateral line neuromast cell expressing GFP. Closed arrow represents a neutrophil expressing DsRED. B, Fluorescent images of zebrafish larvae injected with control dialysate (>14 kDa) or HIC dialysate (>14 kDa) and incubated in 1 μM copper sulfate. C, Comparison of neutrophil count at 1.5 hours after 1-μM copper sulfate treatment in larvae injected with either vehicle, control, or HIC dialysate (>14 kDa). n = 7–8 per group. transport to hospital. In this regard, the feasibility of this approach in humans was recently illustrated in 2 reports demonstrating that remote ischemic conditioning, performed in patients with presumed myocardial infarction by first responders during ambulance transfer to hospital, was able to lessen major cardiac events in patients who subsequently underwent primary percutaneous intervention.12,18 Remote ischemic conditioning, applied as a pre-, per-, or postconditioning stimulus, significantly reduced both the liver inflammation and organ injury caused by S/R. In the context of ischemia/reperfusion injury, neutrophil activation and infiltration into the target organ are considered to play a causal role in mediating injury.17 Several investigators have shown that remote ischemic C 2014 Wolters Kluwer Health, Inc. All rights reserved.
conditioning is able to alter neutrophil activation when studied in the in vitro setting. Shimizu and colleagues19 observed reduced adhesion of neutrophils to tissue culture wells coated with fetal bovine serum when cells were recovered from human volunteers after 3 cycles of 5-minute forearm ischemia/reperfusion. Similarly, transient forearm ischemia/reperfusion was shown to inhibit the rise in CD11b, a critical neutrophil adhesion molecule, induced by subsequent induction of ischemia/reperfusion injury in the same arm.20 Consistent with this finding, Konstantinov et al21 reported that neutrophils recovered from humans subjected to a comparable preconditioning stimulus also expressed reduced levels of surface CD11b. Furthermore, using microarray studies of leukocytes, Konstantinov et al21 demonstrated www.annalsofsurgery.com | 1223
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
the ability of RIPC to downregulate the expression of genes relevant to chemotaxis, adhesion and migration, exocytosis, and innate immunity signaling pathways in neutrophils. These findings of altered neutrophil function in vitro in cells recovered following RIPC protocols suggest a potential mechanism for the protective effect of this intervention. Our experiments demonstrate that the plasma of mice exposed to the HIC protocol can be transferred into the zebrafish and prevent neutrophil migration in the in vivo setting. The zebrafish seems to be an excellent model for in vivo study of neutrophil function, as it shares a number of characteristics with mammalian neutrophils, including the ability to ingest and kill bacteria as well as migratory properties with dynamics that are comparable with those observed in mammalian cells.10 In the present studies, the precise mechanism whereby the transferred plasma impaired leukocyte migration in the zebrafish was not investigated. One recent report suggests that generation of hydrogen peroxide (H2 O2 ) in the region of the tailfin transection is essential for neutrophil migration to this site.22 RIPC has been shown to reduce the production of reactive oxygen species after ischemia/reperfusion injury by upregulation of antioxidative stress genes23,24 and downregulation of proinflammatory genes.25 Therefore, it is possible that the reduction in reactive oxygen species production at the site of injury may contribute to the decreased migration toward the injured site. In their studies, Jan and colleagues9 observed that RIPC was able to augment heme oxygenase activity in the lung and provided evidence that this contributed to the observed beneficial effect. Because heme oxygenase has antioxidative properties and has been shown to neutralize H2 O2 in the in vitro setting,26 this may represent a potential mechanism for the inhibitory effect of ischemic conditioned plasma. The use of a transgenic zebrafish expressing a radiometric H2 O2 sensor gene (HyPer) and also a more detailed investigation of neutrophil directionality in our model may help understand the mechanism underlying the inhibitory effect of ischemic conditioned plasma on neutrophil migration in vivo.22 We determined that the active factor in the transferred plasma had a molecular mass exceeding 14 kDa. With this molecular mass cutoff, multiple proteins could be considered potential candidates for the active molecule. Using a model of liver ischemia/reperfusion, our group previously demonstrated a role for High Mobility Group Box 1 (molecular mass ∼29 kDa) in mediating the protective effect of RIPC.27 Izuishi et al28 also previously showed that exogenously administered High Mobility Group Box 1 prevented liver ischemia/reperfusion injury in a Toll-like receptor 4-dependent. Another obvious candidate is TNF-α, which is also hepatoprotective when administered before liver ischemia/reperfusion.27 Two groups have studied the changes in the plasma proteome after ischemic conditioning in humans and rodents as a means of determining the protective molecule(s).29,30 One interesting novel candidate molecule is apolipoprotein A-1, a 27-kDa protein, which was shown to be increased by ischemic conditioning in both reports. Hibert and colleagues30 further demonstrated that this protein decreased myocardial infarction size when administered to animals before myocardial ischemia/reperfusion. We studied apolipoprotein A-1 levels in control and preconditioned plasma samples and could not detect a significant difference between the 2 groups (data not shown). Shimizu et al16 reported that low molecular mass molecules (
Remote Ischemic Conditioning Prevents Lung and Liver Injury After Hemorrhagic Shock/Resuscitation Potential Role of a Humoral Plasma Factor Chung Ho Leung, BSc,∗ †‡ Christopher A. Caldarone, MD,†‡ Feng Wang, MSc,∗ Seetha Venkateswaran, BSc,∗ Menachem Ailenberg, PhD,∗ Brian Vadasz, BSc,∗ ‡ Xiao-Yan Wen, PhD,∗ ‡ and Ori D. Rotstein, MD∗ ‡ Objective: To evaluate the efficacy of remote ischemic conditioning (RIC) on organ protection after hemorrhagic shock/resuscitation (S/R) in a murine model. Background: Ischemia/reperfusion resulting from S/R contributes to multiple organ dysfunction in trauma patients. We hypothesized that RIC before shock (remote ischemic preconditioning), during shock (remote ischemic “PER”conditioning), or during resuscitation (remote ischemic “POST”conditioning) could confer organ protection. We also tested the effect of ischemic conditioned plasma on neutrophil migration in vivo using transgenic zebrafish models. Methods: C57Bl/6 mice were subjected to S/R with or without hindlimb RIC. Serum levels of alanine aminotransferase and tumor necrosis factor-alpha, and liver tumor necrosis factor-alpha and interleukin 1β mRNA were evaluated. In some experiments, lung protein leakage, cytokines, and myeloperoxidase activity were investigated. Plasma from mice subjected to RIC was microinjected into zebrafish, and neutrophil migration was assessed after tailfin transection or copper sulfate treatment. Results: In mice subjected to S/R, remote ischemic preconditioning, remote ischemic “PER”conditioning, and remote ischemic “POST”conditioning each significantly reduced serum alanine aminotransferase and liver mRNA expression of tumor necrosis factor-alpha and interleukin 1β and improved liver histology compared with control S/R mice. Lung injury and inflammation were also significantly reduced in mice treated with remote ischemic preconditioning. Zebrafish injected with plasma or dialyzed plasma (fraction >14 kDa) from ischemic conditioned mice had reduced neutrophil migration toward sites of injury compared with zebrafish injected with control plasma. Conclusions: RIC protects against S/R-induced organ injury, in part, through a humoral factor(s), which alters neutrophil function. The beneficial effects of RIC, performed during the S/R phase of care, suggest a role for its application early in the posttrauma period. Keywords: hemorrhagic shock, ischemia/reperfusion, remote ischemic conditioning, resuscitation, trauma (Ann Surg 2015;261:1215–1225)
H
emorrhagic shock and resuscitation (S/R) after traumatic injury contributes to the development of the systemic inflammatory response and multiorgan dysfunction in this patient population.1 Several mechanisms have been invoked to explain this response to From the ∗ Department of Surgery, the Keenan Research Centre for Biomedical Science and the Zebrafish Centre for Advanced Drug Discovery, St. Michael’s Hospital, Toronto, Ontario, Canada; †Division of Cardiovascular Surgery, Hospital for Sick Children, Toronto, Ontario, Canada; and ‡Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada. Disclosure: Supported by the Canadian Institutes of Health Research grant #37779. The authors declare no conflicts of interest. Reprints: Ori D. Rotstein, MD, St. Michael’s Hospital, 30 Bond St 16CC-044, Toronto, ON M5B 1W8, Canada. E-mail: [email protected]. C 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright ISSN: 0003-4932/14/26106-1215 DOI: 10.1097/SLA.0000000000000877
Annals of Surgery r Volume 261, Number 6, June 2015
S/R. One unifying theme is that S/R represents a form of global ischemia/reperfusion, which initiates a broad range of cellular signaling events, which directly or indirectly promote cellular injury and inflammation. For example, oxidative stress derived from ischemia/reperfusion not only contributes directly to local organ injury through the generation of injurious reactive oxygen species but also participates in cell signaling events leading to the activation of inflammatory molecules.2,3 In addition, S/R primes immune cells to exhibit an exaggerated phenotype, leading to subsequent lung injury in a second hit.4 Strategies directed at lessening ischemia/reperfusion injury caused by S/R have potential to improve outcome in this patient population. One potent intervention that harnesses the cell’s innate protective mechanism against ischemia/reperfusion injury is known as remote ischemic preconditioning (RIPC, reviewed by Szij´art´o et al5 ). In experimental RIPC in animals, cycles of transient ischemia/reperfusion applied to the limb have been shown to reduce subsequent ischemia/reperfusion injury in multiple distant organs including the heart, liver, lungs, and brain.5 In humans, RIPC also seems to exert beneficial effects. Its effects have been most thoroughly studied in the context of diminishing myocardial injury after surgical or radiological interventions.6 The concept of ischemic conditioning was initially described in reference to a conditioning stimulus applied before ischemia/reperfusion of the target organ (ie, “preconditioning”).7 More recently, the concept of ischemic conditioning has been extended to provide the conditioning stimulus during ischemia of the target organ (“perconditioning”) and after ischemia of the target organ (“postconditioning”). The extension of effective organ conditioning to include these periods broadens the potential clinical use of this protective intervention. Like preconditioning, both per- and postconditioning have received the greatest attention in the cardiac domain, although studies have demonstrated the efficacy of these approaches to ischemia/reperfusion injury in other organs.5,8 In the context of S/R, preconditioning may have potential relevance to the military setting, where combatants at risk of sustaining later injury may submit to RIPC before engaging in combat. In this regard, a single study recently reported by Jan and colleagues9 demonstrated the ability of RIPC to mitigate shock-induced lung injury in a rat model. However, demonstration of effective per- and postconditioning would further extend its use in the trauma setting. Specifically, should these latter strategies prove efficacious, this would open the window for medical personnel to use remote ischemic conditioning during the shock or resuscitation phases of trauma care, with a view to lessening the consequences of S/R. The present studies evaluate the ability of remote ischemic conditioning to reduce liver and lung injury in a murine model of hemorrhagic shock and to investigate potential mechanisms for protection. We show that preconditioning, as well as perconditioning and postconditioning, are each able to mitigate organ injury and lessen inflammation after S/R. Furthermore, we investigated a potential mechanism underlying the beneficial effect of remote ischemic www.annalsofsurgery.com | 1215
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
conditioning. Using the zebrafish to study wound-induced neutrophil migration in vivo,10 we demonstrate that a humoral plasma factor(s) generated by the ischemic conditioning protocol in mice inhibits neutrophil migration to sites of injury.
(A)
Resuscitaon (2-volume saline of shed blood)
Hemorrhagic Shock (22.5 μl blood/g body weight)
Control S/R
METHODS
60’
Animal Models of Hemorrhagic S/R In preliminary studies, we examined the effect of different periods of hemorrhagic shock on the development of liver injury in the reperfusion phase. Consistent with the literature,11 the duration of hypovolemic shock was an important determinant of the magnitude of liver injury. We observed that 1 hour of shock induced mild liver injury whereas a 2-hour hypotensive period produced much more severe injury with alanine aminotransferase (ALT) levels that were approximately 10-fold higher than those with 1 hour of hypotension (data not shown). This afforded us the opportunity to study the effect of ischemic conditioning in models of both mild and severe injuries.
Remote Ischemic Conditioning Protocol in the S/R Model Inducing Mild Liver Injury Adult male C57Bl/6 mice (Charles River Laboratories, St Constant, Quebec), 9 to 11 weeks old, weighing 22 to 28 g, were anesthetized with isoflurane and the right carotid artery was cannulated by a PE-10 microtube for hemorrhagic S/R. Before the initiation of hemorrhagic shock, a cycle of left femoral artery occlusion for 10 minutes, followed by reperfusion for 10 minutes, was performed as a means of inducing RIPC. Control S/R animals had undergone groin dissection and isolation of the femoral artery but without artery occlusion. Hemorrhagic shock was accomplished by a 1-time blood withdrawal (22.5 μL of blood/g body weight) over 15 minutes. The amount of blood withdrawn lowered the mean arterial blood pressure to 30 mm Hg as monitored by a transducer via the carotid cannula. After a hypotensive period of 60 minutes, animals were resuscitated with 0.9% normal saline equivalent to twice the volume of blood withdrawn over 15 minutes. At 1 and 2 hours after resuscitation, liver and lung tissues and blood samples were collected for analysis (Fig. 1A).
Remote Ischemic Conditioning Protocol in the S/R Model Inducing Severe Liver Injury Mice were subjected to hemorrhagic shock at 30 mm Hg of mean arterial blood pressure for 2 hours by blood withdrawal from the left femoral artery. Mean arterial blood pressure was continuously monitored through the left carotid artery over this period. Mice were then resuscitated with shed blood and 2 volumes of lactated Ringer’s solution over 15 minutes. At 2 hours after resuscitation, liver tissue and blood samples were collected for analysis. Using this model, the effect of remote ischemic conditioning on liver injury was investigated. We modified these protocols to more closely mimic that reported both in experimental animals and in humans.8,12 Specifically, RIPC was accomplished by performing 4 cycles of alternating 5-minute ischemia, followed by 5-minute reperfusion of the right lower extremity, using an externally applied tourniquet before shock. Control S/R animals were subjected to anesthesia but without hindlimb ischemia. Given the fact that, in the trauma setting, it would be unlikely to be able to perform preconditioning, we also examined the effect of remote ischemic “PER”conditioning (RIPerC) as well as combined RIPerC and “POST”conditioning (RIPostC) on liver injury. In RIPerC, the 4 cycles are performed during the final 40 minutes of the shock period, whereas in the RIPerC + RIPostC, the first of the 4 cycles is performed during the last 10 minutes of shock and the final 3 cycles are carried into the resuscitation period (Fig. 1B). 1216 | www.annalsofsurgery.com
15’
120’
15’
120’
Remote Ischemic Precondioning (RIPC) 10
60’
10
Hemorrhagic Shock (30 mm HG MAP)
(B)
Resuscitaon (Shed Blood + 2-Volume Ringer’s Lactate)
Control S/R
120’
15’
120’
15’
120’
15’
120’
Remote Ischemic Precondioning (RIPC)
120’
55
Remote Ischemic Percondioning (RIPerC)
120’
55
RIPerC + Remote Ischemic Postcondioning (RIPerC+RIPostC)
120’
15’
120’
55
FIGURE 1. A, Schematic of hemorrhagic S/R protocol of mild liver injury. Remote ischemic preconditioning was performed by 10-minute femoral artery occlusion, followed by 10-minute reperfusion. B, Schematic of hemorrhagic S/R protocol of severe liver injury. Remote ischemic conditioning was performed by 4 cycles of 5-minute ischemia and 5-minute reperfusion by a tourniquet.
Serum ALT ALT levels as a marker of hepatocellular injury were measured in blood samples taken at 1 and 2 hours after resuscitation by the Diagnostic Laboratory at St. Michael’s Hospital.
Liver Histology Liver tissue obtained at 2 hours after resuscitation was fixed by immersion in formaldehyde, embedded by paraffin wax, and cut in 5-μm slices. Sections were stained with hematoxylin and eosin and evaluated by light microscopy in a blinded manner by an independent person using the following scoring criteria13 : 0, no hepatocellular damage; 1, mild injury characterized by cytoplasmic vacuolization and focal nuclear pyknosis; 2, moderate injury with dilated sinusoids, cytosolic vacuolization, and blurring of intercellular borders; 3, moderate-to-severe injury with coagulative necrosis, abundant sinusoidal dilation, red blood cell extravasation into hepatic chords, and hypereosinophilia and margination of neutrophils; 4, severe necrosis with loss of hepatic architecture, disintegration of hepatic cords, hemorrhage, and neutrophil infiltration. C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Limb Conditioning Prevents Organ Injury
Real-time Quantitative Reverse Transcription Polymerase Chain Reaction of Inflammatory Gene mRNA Expression
GCCAAGTATGATGACA and TGAAGTCGCAGGAGACAACCT, respectively.
TNF-α Enzyme-linked Immunosorbent Assay
Total RNA from the liver and lung were extracted using RNeasy Mini Kit (Qiagen, Valencia, CA). One microgram of RNA was then subjected to DNAse treatment and reverse transcription (Bio-rad, Hercules, CA). The relative abundance of mRNA for tumor necrosis factor alpha (TNF-α), interleukin 1β (IL-1β), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, endogenous control) in the liver and lung tissues were quantified using SYBR green-based quantitative polymerase chain reaction (PCR) (Applied Biosystems, Foster City, CA). Each reaction was performed in triplicate. Realtime PCR conditions were as follows: initial denaturation at 95◦ C for 10 minutes, followed by 40 cycles of 95◦ C for 15 seconds, and 60◦ C for 60 seconds. Purity of the PCR reaction was verified by the meltcurve analysis at the completion of the PCR. Relative abundance of gene expression was calculated by the 2−CT method, using GAPDH as the endogenous control and one of the sham-operated samples as reference. All primers were purchased from Sigma-Aldrich, Oakville, Ontario, Canada. The sequences of forward and reverse primers for TNF-α are TATGGCTCAGGGTCCAACTC and CTCCCTTTGCAGAACTCAGG, respectively. The sequences of forward and reverse primers for IL-1β are GCCCATCCTCTGTGACTCAT and AGGCCACAGGTATTTTGTCG, respectively. The sequences of forward and reverse primers for GAPDH are AGAAACCT-
Serum levels of TNF-α were quantitated in blood samples taken at 1 and 2 hours into resuscitation using commercially available enzyme-linked immunosorbent assay kits according to the manufacturer’s instructions (R&D Systems, Minneapolis, MN).
TNF-α Western Blots Liver and lung tissue protein homogenates were separated via 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Bio-Rad Laboratories, Inc., Hercules, CA). The membranes were then blocked with 5% nonfat milk and probed with TNF-α rat antibody (BioLegend, San Diego, CA) and GAPDH rabbit antibody (Cell Signaling, Danvers, MA) for loading control, followed by appropriate secondary antibodies conjugated with horseradish peroxidase. The blots were exposed to film after enhanced chemiluminescence (Amersham Biosciences, Piscataway, NJ). Densitometry of the blots was analyzed by Quantity One (Bio-Rad Laboratories, Inc., Hercules, CA).
Bronchoalveolar Lavage At the end of resuscitation, tracheostomy was performed and the lungs were lavaged 3 times with cold phosphate-buffered saline in
(A)
(B)
80
Serum ALT (U/L)
70
P < 0.01
Liver TNF-α mRNA (Fold over Sham)
P < 0.05
90
P < 0.01
60 50 40 30 20 10
8
P < 0.05
7 6 5 4 3 2 1 0
0 Sham Operated Control S/R 1 h
RIPC S/R 1 h
Control S/R 2 h
P < 0.01 (C)
Sham operated
RIPC S/R 2 h
Control S/R 2 h
RIPC S/R 2 h
P < 0.05
Liver IL-1β mRNA (Fold over Sham)
30 25 20 15 10 5 0 Sham operated
Control S/R 2 h
RIPC S/R 2 h
FIGURE 2. A, In the mild liver injury model, S/R resulted in liver injury as indicated by increased serum ALT levels at 1 and 2 hours after resuscitation in control S/R mice. Serum ALT was significantly reduced in mice with RIPC. n = 5–8 per group. B, mRNA expression of TNF-α evaluated by quantitative real-time-PCR in the liver after S/R. n = 4–5 per group. C, IL-1β mRNA expression in the liver after S/R. n = 4–5 per group. C 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 1217
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
(B)
3
Liver Histology Injury Score
(A)
2.5
P < 0.01
2 1.5 1 0.5 0 Control S/R 2 h
RIPC S/R 2 h
FIGURE 3. A, Liver histology imaged at 40× magnification by hematoxylin and eosin staining after S/R. Compared with shamoperated mice, S/R significantly increased overall hepatic ballooning degeneration and loss of sinusoidal architecture. However, these were reduced in mice with RIPC. B, Liver injury histology score is significantly lower in mice treated with RIPC. n = 5 per group.
800-μL aliquots. Total protein content in the bronchoalveolar lavage fluid (BAL) fluid as an indicator of lung injury was assayed using a protein assay kit (Bio-Rad Laboratories, Inc., Hercules, CA).
Lung tissue was homogenized and sonicated in 0.5% hexadecyltrimethylammonium bromide (Sigma, Oakville, Ontario, Canada) in 50 mM potassium phosphate buffer (pH 6.0) and assayed with 0.167 mg/mL of o-dianisidine (Sigma, Oakville, Ontario, Canada) and 0.005% hydrogen peroxide. The change in absorbance was measured at 460 nm (ε = 1.3 × 104 M−1 ·cm−1 ). Myeloperoxidase (MPO) activity is expressed as units of MPO per milligram of lung protein, whereby 1 unit of MPO is defined as the amount of enzyme degrading 1 nmol hydrogen peroxide per minute.
P < 0.01
P < 0.01 P < 0.05
600 Serum TNF-α (pg/mL)
Myeloperoxidase Enzyme Activity
700
500 400 300 200 100 0 Sham Operated Control S/R 1 h
RIPC S/R 1 h
Control S/R 2 h
RIPC S/R 2 h
FIGURE 4. Serum levels of TNF-α were significantly increased after S/R and this was significantly prevented in mice with RIPC. n = 3–5 per group.
Zebrafish Models of In Vivo Neutrophil Infiltration We used a transgenic zebrafish [Tg(mpx::GFP)] model with specific green fluorescent protein (GFP)-tagged neutrophils (GFP controlled by the MPO gene promoter) to monitor, in real time, the migration of neutrophils to the wounded area in response to injury induced by tailfin transection.14 We also used a chemical injury model in a separate transgenic zebrafish line (cldnB::GFP/lysC::DsRED2) by copper sulfate treatment, which induces an inflammatory process characterized by migration of neutrophils, labeled with red fluorescent protein (DsRED), to the copper sulfate-injured neuromast cells (labeled with GFP).15 1218 | www.annalsofsurgery.com
Preparation of Remote Ischemic Conditioning Plasma for Zebrafish Microinjection For these studies, we obtained blood from the mice after 4 cycles of hindlimb ischemic conditioning (HIC) to study the effect of plasma on neutrophil function using zebrafish models of neutrophil migration. Blood was collected at 15 minutes after the last cycle of conditioning into sodium citrate tubes for plasma collection. To determine a molecular weight cutoff for humoral factor(s) responsible for conditioning, control and HIC plasma were dialyzed in a C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Limb Conditioning Prevents Organ Injury
(A)
(B)
P < 0.01
P < 0.01 0.8 0.7
P < 0.01
P < 0.01
P < 0.01
P < 0.05
90
0.6
MPO U/mg lung proen
BAL Protein (mg/mL)
80 0.5 0.4 0.3 0.2 0.1
70 60 50 40 30 20 10
0 Sham Operated Control S/R 1 h
RIPC S/R 1 h
Control S/R 2 h
RIPC S/R 2 h
0 Sham Operated
P < 0.01
(C)
(D)
P < 0.01
P < 0.01
20 Lung IL-1β mRNA (Fold over Sham)
Lung TNF-α mRNA (Fold over Sham)
4 3.5 3 2.5 2 1.5 1 0.5
Control S/R 2 h
RIPC S/R 2 h
P < 0.01
18 16 14 12 10 8 6 4 2 0
0
Sham operated
Sham operated
Control S/R 2 h
Control S/R 2 h
RIPC S/R 2 h
RIPC S/R 2 h
(E) Sham Operated
Control S/R 2 h
RIPC S/R 2h
TNF-α
GAPDH
3
P < 0.05
P < 0.01
Lung TNF-α/GAPDH
2.5 2 1.5 1 0.5 0 Sham Operated
Control S/R 2 h
RIPC S/R 2 h
FIGURE 5. A, Lung injury as indicated by protein leakage in the BAL is significantly increased in control mice after S/R. RIPC prevented this increase. n = 3–5 per group. B, MPO enzymatic activity in lung tissue as an indicator of neutrophil infiltration is significantly increased at 2 hours after resuscitation in control mice. RIPC prevented this increase. n = 4–6 per group. C, TNF-α mRNA expression in the lung after S/R. D, IL-1β mRNA expression in the lung after S/R. n = 5 per group. E, Western blot analysis of TNF-α protein expression in lung tissue at 2 hours after resuscitation. n = 3 per group.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 1219
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
14-kDa cutoff dialysis membrane (Spectrum, Rancho Dominguez, CA) overnight.16
(A) 900 800
cldnB::GFP and lysC::DsRED2 Zebrafish: Chemical Injury by Copper Sulfate cldnB::GFP zebrafish expressing GFP on neuromast cells were crossed with lysC::DsRED2 zebrafish expressing DsRED on neutrophils to produce larvae expressing both fluorescent proteins. Fourday postfertilization larvae were injected with control or HIC dialysate (>14 kDa) and were incubated for 24 hours before treatment with copper sulfate (1 μM) for 1.5 hours (peak neutrophil count). Neutrophil migration localized to the lateral line neuromast cell areas was quantified.
Serum ALT (U/L)
Three days postfertilization Tg(mpx::GFP) larvae were anesthetized by immersion in clove oil. To evaluate the effect of HIC on neutrophils, larvae were injected with vehicle (saline), control, or HIC plasma (1:20 dilution in saline), or the dialyzed fractions (>14 kDa and 14 kDa) significantly reduced neutrophil counts in the wounded area compared with the control dialysate (>14 kDa) (P < 0.004). In contrast, neutrophil counts after injection with HIC dialysate (14 kDa
HIC Dialysate >14 kDa
FIGURE 9. A, Fluorescent image of an untreated cldnB::GFP/lysC::DsRED2 zebrafish. Open arrow represents a lateral line neuromast cell expressing GFP. Closed arrow represents a neutrophil expressing DsRED. B, Fluorescent images of zebrafish larvae injected with control dialysate (>14 kDa) or HIC dialysate (>14 kDa) and incubated in 1 μM copper sulfate. C, Comparison of neutrophil count at 1.5 hours after 1-μM copper sulfate treatment in larvae injected with either vehicle, control, or HIC dialysate (>14 kDa). n = 7–8 per group. transport to hospital. In this regard, the feasibility of this approach in humans was recently illustrated in 2 reports demonstrating that remote ischemic conditioning, performed in patients with presumed myocardial infarction by first responders during ambulance transfer to hospital, was able to lessen major cardiac events in patients who subsequently underwent primary percutaneous intervention.12,18 Remote ischemic conditioning, applied as a pre-, per-, or postconditioning stimulus, significantly reduced both the liver inflammation and organ injury caused by S/R. In the context of ischemia/reperfusion injury, neutrophil activation and infiltration into the target organ are considered to play a causal role in mediating injury.17 Several investigators have shown that remote ischemic C 2014 Wolters Kluwer Health, Inc. All rights reserved.
conditioning is able to alter neutrophil activation when studied in the in vitro setting. Shimizu and colleagues19 observed reduced adhesion of neutrophils to tissue culture wells coated with fetal bovine serum when cells were recovered from human volunteers after 3 cycles of 5-minute forearm ischemia/reperfusion. Similarly, transient forearm ischemia/reperfusion was shown to inhibit the rise in CD11b, a critical neutrophil adhesion molecule, induced by subsequent induction of ischemia/reperfusion injury in the same arm.20 Consistent with this finding, Konstantinov et al21 reported that neutrophils recovered from humans subjected to a comparable preconditioning stimulus also expressed reduced levels of surface CD11b. Furthermore, using microarray studies of leukocytes, Konstantinov et al21 demonstrated www.annalsofsurgery.com | 1223
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 6, June 2015
Leung et al
the ability of RIPC to downregulate the expression of genes relevant to chemotaxis, adhesion and migration, exocytosis, and innate immunity signaling pathways in neutrophils. These findings of altered neutrophil function in vitro in cells recovered following RIPC protocols suggest a potential mechanism for the protective effect of this intervention. Our experiments demonstrate that the plasma of mice exposed to the HIC protocol can be transferred into the zebrafish and prevent neutrophil migration in the in vivo setting. The zebrafish seems to be an excellent model for in vivo study of neutrophil function, as it shares a number of characteristics with mammalian neutrophils, including the ability to ingest and kill bacteria as well as migratory properties with dynamics that are comparable with those observed in mammalian cells.10 In the present studies, the precise mechanism whereby the transferred plasma impaired leukocyte migration in the zebrafish was not investigated. One recent report suggests that generation of hydrogen peroxide (H2 O2 ) in the region of the tailfin transection is essential for neutrophil migration to this site.22 RIPC has been shown to reduce the production of reactive oxygen species after ischemia/reperfusion injury by upregulation of antioxidative stress genes23,24 and downregulation of proinflammatory genes.25 Therefore, it is possible that the reduction in reactive oxygen species production at the site of injury may contribute to the decreased migration toward the injured site. In their studies, Jan and colleagues9 observed that RIPC was able to augment heme oxygenase activity in the lung and provided evidence that this contributed to the observed beneficial effect. Because heme oxygenase has antioxidative properties and has been shown to neutralize H2 O2 in the in vitro setting,26 this may represent a potential mechanism for the inhibitory effect of ischemic conditioned plasma. The use of a transgenic zebrafish expressing a radiometric H2 O2 sensor gene (HyPer) and also a more detailed investigation of neutrophil directionality in our model may help understand the mechanism underlying the inhibitory effect of ischemic conditioned plasma on neutrophil migration in vivo.22 We determined that the active factor in the transferred plasma had a molecular mass exceeding 14 kDa. With this molecular mass cutoff, multiple proteins could be considered potential candidates for the active molecule. Using a model of liver ischemia/reperfusion, our group previously demonstrated a role for High Mobility Group Box 1 (molecular mass ∼29 kDa) in mediating the protective effect of RIPC.27 Izuishi et al28 also previously showed that exogenously administered High Mobility Group Box 1 prevented liver ischemia/reperfusion injury in a Toll-like receptor 4-dependent. Another obvious candidate is TNF-α, which is also hepatoprotective when administered before liver ischemia/reperfusion.27 Two groups have studied the changes in the plasma proteome after ischemic conditioning in humans and rodents as a means of determining the protective molecule(s).29,30 One interesting novel candidate molecule is apolipoprotein A-1, a 27-kDa protein, which was shown to be increased by ischemic conditioning in both reports. Hibert and colleagues30 further demonstrated that this protein decreased myocardial infarction size when administered to animals before myocardial ischemia/reperfusion. We studied apolipoprotein A-1 levels in control and preconditioned plasma samples and could not detect a significant difference between the 2 groups (data not shown). Shimizu et al16 reported that low molecular mass molecules (
