|
|
||||||||
J Thorac Cardiovasc Surg 2007;133:1588-1596
© 2007 The American Association for Thoracic Surgery
Cardiopulmonary Support and Physiology |
a Department of Pediatric Cardiac Surgery, Oregon Health Sciences University, Portland, Ore
b Department of Neuropathology, Oregon Health Sciences University, Portland, Ore
c Department of Immunology, Oregon Health Sciences University, Portland, Ore.
Presented as a poster presentation at the Thirty-second Annual Meeting of the Western Thoracic Surgical Association, Sun Valley, Idaho, June 21-24, 2006.
Received for publication July 27, 2006; revisions received November 15, 2006; accepted for publication December 18, 2006. * Address for reprints: Edward J. Hickey, MD, CHSS Data Center, Hospital for Sick Children, 555 University Ave, Toronto, Ontario M5G 1X8, Canada. (Email: hickeydoc{at}yahoo.com).
| Abstract |
|---|
|
|
|---|
Methods: Neonatal piglets were blindly and randomly preconditioned with lipopolysaccharide (n = 6) or saline (n = 6). Three days later, they experienced 2 hours of deep hypothermic circulatory arrest before being weaned and supported anesthetized for 20 hours in an intensive care setting. Controls included cardiopulmonary bypass without deep hypothermic circulatory arrest (n = 3) and no cardiopulmonary bypass (n = 3). Brain injury was quantified by light and fluorescent microscopy (Fluoro-Jade; Histo-Chem, Inc, Jefferson, Ark).
Results: All animals were clinically indistinguishable before surgery. Perioperative and postoperative parameters between experimental groups were similar. No control animal scored falsely positive. Histologic scores were 0.33 ± 0.21, 0.66 ± 0.42, and 0.5 ± 0.24 in the cortex, basal ganglia, and hippocampus, respectively, in the lipopolysaccharide-treated animals but significantly worse in all saline control animals (1.33 ± 0.21, P < .01; 1.66 ± 0.33, P = .09; and 6.0 ± 1.5, P < .01). One lipopolysaccharide-treated brain was histologically indistinguishable from controls.
Conclusion: This is the first evidence that lipopolysaccharide can precondition against cardiopulmonary bypassrelated injury. Because lipopolysaccharide preconditioning is a systemic phenomenon offering proven protection against myocardial, hepatic, and pulmonary injury, this technique offers enormous potential for protecting against systemic neonatal injury related to cardiopulmonary bypass.
| Introduction |
|---|
|
|
|---|
A widely recognized trigger for delayed preconditioning is bacterial lipopolysaccharide (LPS) and its derivatives. Systemically administered in extremely low doses, it provides protection against a variety of subsequent insults, including myocardial, renal, liver, and cerebral ischemia, where infarct volume can be reduced by as much as 60%.2
Evidence has recently emerged suggesting that delayed preconditioning involves a complete reprogramming at the genetic level, to provide systemic protection against both ischemic injury3
and activation of components of the cellular immune response.4
This concept, therefore, is extremely appealing in the context of ischemic CPB-related injury, where systemic inflammation, hypoxiaischemia, and reperfusion injury act synergistically to cause end-organ damage.
Because LPS preconditioning consistently induces robust protection against ischemic brain injury in experimental rodents, we entertained the hypothesis that low-dose LPS may confer cerebral protection against ischemic injury sustained during periods of neonatal deep hypothermic circulatory arrest (DHCA). Using a piglet model of DHCA, involving a 20-hour reperfusion period in a fully monitored intensive care environment, we aimed to test this hypothesis by histologically quantifying regional brain injury after animals were blindly and randomly inoculated before surgery with either a preconditioning dose of LPS or saline placebo.
| Materials and Methods |
|---|
|
|
|---|
Surgical Procedures
All animal experiments were conducted with the approval of the institutions Animal Care and Use Committee. The animals received humane care in compliance with the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health (NIH publication No. 85-23, revised 1995).
The surgical protocol, including CPB circuit and prime, was similar to that described previously by our laboratory.5
In brief, after anesthetic induction (1% isofluorane) and placement of peripheral lines and temperature probes, surgical tracheotomy was performed to allow controlled ventilation, maintaining arterial oxygen and carbon dioxide tensions within normal limits.
The heart was exposed via a median sternotomy. After systemic heparinization (500 IU/kg), the aortic root and right atrial appendage were cannulated and CPB was established at 100 to 150 mL · kg1 · min1 to maintain mean arterial pressures of 50 mm Hg. The animal was then perfusion-cooled with a pH-stat strategy to 18°C over 30 minutes. CPB was then ceased and the circulatory volume drained into the reservoir to induce uninterrupted circulatory arrest for 120 minutes. During this period, myocardial protection was afforded by slowly perfusing through an isolate coronary circulation. An aortic crossclamp was applied to the ascending aorta immediately distal to the aortic cannula, and soft clips were applied to the inferior and superior venae cavae, thereby isolating the coronary circulation through which cold blood was administered at a rate of 5 to 10 mL/min. Full-flow CPB was later re-established and the animal rewarmed, with the administration of carbonate (8.4%) to correct acidosis. The lungs were ventilated and the animals separated from CPB in conjunction with a continuous infusion of intravenous norepinephrine (2 mL/h) and volume replacement as necessary.
Postoperative Care
Protamine was administered (5 mg/1000 IU heparin), drains and bladder catheters were positioned, and all wounds were closed. Animals were then supported in a laboratory intensive care, anesthetized (intravenous propofol 3-5 mL/h), and ventilated, with fully alarmed invasive monitoring and continuous veterinary support. The arterial pressure was maintained with a mean greater than 50 mm Hg. After 20 hours of reperfusion, the animals were exsanguinated, irrigated (3 L of saline), and perfusion-fixed (2 L of ice-cold 4% paraformaldehyde). The brains were retrieved intact before being sectioned coronally, embedded in paraffin blocks, and 5-µm slices mounted on slides for staining with hematoxylin and eosin or Fluoro-Jade.
Staining with Fluoro-Jade
Fluoro-Jade is an anionic fluorochrome that selectively stains irreversibly injured neurons with greater sensitivity and specificity than hematoxylin and eosin.6
After deparaffinization, slides were immersed for 20 minutes in 0.06% potassium permanganate before being washed several times in deionized water. Fluoro-Jade crystals were dissolved to a 0.0004% solution in 0.01% acetic acid and the slides were immersed for 20 minutes before thorough air-drying in a warm incubator. After xylene washes, slides were covered and viewed with an epifluorescent microscope with blue (450-490 nm) excitation light. A barrier filter allowing passage of all wavelengths longer than 515 nm results in a green emission color from cells stained positively with Fluoro-Jade.
Histologic Scoring
Slides were examined randomly by investigators blinded to the experimental protocol. Morphologic criteria used to identify ischemic neurons were a prominent eosinophilic nucleus, with lack of a well-defined nucleolus and Nissl substance, together with nuclear fragmentation and formation of dark apoptotic bodies.7
These are established signs of irreversible neuronal injury and avoid false-positive identification of hyperchromatic neuronal artifact.7
Qualitative grading scores were applied according to the cerebral cortex, basal ganglia, hippocampus, and cerebellum (where 0 = normal and 4 = confluent, severe injury). The hippocampus is know to exhibit regional vulnerability, and therefore subregions (CA1, CA2, CA3, CA4, and superficial, middle, and deep dentate gyrus) were selectively scored, inasmuch as repeated studies with our model have demonstrated reproducible hierarchical injury within these areas.
Statistical Analysis
All data are expressed as mean ± standard error of the mean. The unpaired Student t tests were used to compare perioperative parameters and quantitative histologic scores between animals in the saline and LPS preconditioned groups or otherwise analysis of variance to compare values between groups with SAS statistical software (SAS, Inc, Cary, NC). Statistical significance was tested to the 95% confidence limit.
| Results |
|---|
|
|
|---|
|
|
|
|
Fluoroscopic assessment with Fluoro-Jade reiterated the changes observed by light microscopy (Table 4 and Figure 2). Fluoro-Jade is highly sensitive and specific for irreversible neuronal injury and makes identification of positive neurons straightforward (Figure 3). Every brain region examined in this study except the cerebellum exhibited significantly lesser injury by Fluoro-Jade assessment after LPS preconditioning. No false-positive scores were observed in either the nonischemic control or sham animals.
|
|
|
| Discussion |
|---|
|
|
|---|
Cerebral preconditioning is not a new idea. Several acute preconditioning techniques are widely established experimentally, in particular brief hypoxiaischemia (or hypoxia mimetics9
). It is surprising, therefore, that these techniques were tested against CPB-related brain injury only recently.9,10
The initial application to CPB by Yeh and associates,10
involved the use of acute ischemic preconditioning. By contrast, delayed preconditioning is a considerably more appealing concept, not least because it offers a significantly greater window of protection and potency.2
In clinical terms, for example, a 7-day window of tolerance would span the entire operative and intensive care periods, during which numerous compounding factors are believed to influence progression of the initial injury.
Although several agents, including ischemiahypoxia (and their mimetics9
), can induce delayed protection, LPS and its derivatives are among the most closely studied. LPS preconditioning was originally noted to confer protection against subsequent lethal doses of LPS (homologous tolerance).11
However, it also exhibits cross-tolerance, where a nonlethal stress confers tolerance to a subsequent stress that is different in nature from the first. Thus, systemic LPS has now been shown to also induce ischemic tolerance in several systems including the liver, myocardium, kidney, and brain.11-13
In fact, the characteristics of LPS preconditioning appear to share several hallmark features with delayed ischemic preconditioning in the brain,11
suggesting that common pathways may link the numerous primary preconditioning stimuli.
Rodent models generating large infarct zones consistently demonstrate reduction of injury by as much as 60% after systemic LPS inoculation.2
Histologic models of injury after DHCA instead require extended ischemic periods to generate scattered evidence of ischemic injury.9,14,15
These extended durations of DHCA represent a departure from clinical practice and are therefore a significant limitation in the experimental study of DHCA. Nevertheless, whereas all nonpreconditioned animals in this present study displayed brain injury in all regions other than cerebellum, brain regions from some of the preconditioned animals were indistinguishable from control groups, with no ischemic injury apparent, suggesting a robust protective effect. The premise on which exaggerated animal DHCA models are used is that observed protection should translate to some degree of protection against clinical DHCA injury, however slight. In addition, the protection demonstrated in this present study is important for several reasons: (1) LPS preconditioning has never been tested in conjunction with CPB; (2) although descriptions of LPS preconditioning exist in swine,16
none have examined cerebral protection; and (3) conflicting reports exist as to whether or not the neonatal mammalian brain is amenable to preconditioning.17-19
| Conclusions |
|---|
|
|
|---|
LPS is a ligand for Toll-like receptor (TLR)-4.25
TLRs are evolutionary historic transmembrane proteins that recognize generic constituents of foreign bacteria (double-stranded RNA, unmethylated DNA sequences, and LPSs) to trigger downstream inflammatory pathways. In addition, however, it is recognized that injured and necrotic cells also release endogenous ligands, including RNA, chromosomal DNA, heat shock proteins, and components of extracellular matrix.26
TLRs may therefore represent the common pathway linking diverse initiators of delayed preconditioning.
One theory proposed26
is that a very low level of stimulation of TLRs (either through exogenous LPS or endogenous components generated through a preconditioning stimulus such as subinjurious ischemia), together with proinflammatory cytokines (tumor necrosis factor-
[TNF-
], interleukin-1), results in downstream generation of both proinflammatory and anti-inflammatory cytokines. In addition, suppressors of inflammation are produced, which include decoy receptors and decoy ligands. During the second stimulation, these decoys and anti-inflammatory cytokines predominate, resulting in an overall reduction in tissue injury. Alternatively, if the initial preconditioning stimulus exceeds beyond a threshold dose, the initial response is overwhelmed by pro-inflammatory processes, and the preconditioning response will not occur. Thus, preconditioning is similar to a "trip switch." If the preconditioning dose falls above the threshold, the trip is flipped, and tissue injury will occur instead.
The above hypothesis would explain several conundrums, including the paradox as to why an obligatory step in the delayed preconditioning process is the generation of potentially harmful inflammatory mediators. TNF-
in particular, and its downstream signaling mediator ceramide, are critical effectors of LPS-induced neuroprotection.27,28
LPS treatment accompanied by concomitant neutralizing antiTNF-
reverses the protective effect of LPS preconditioning.11
Further downstream, the effector pathways are less clear, but one suggestion that is gathering support involves a central role of interferon-ß, secondary to the TNF-signaling axis (reviewed in reference 1
). A neuroprotective role of interferon-ß is supported by the recent finding that its systemic administration improves stroke outcome.29
The precise cascade of intracellular signaling events is clearly complex. However, the end result appears to be an entire genetic reprogramming of the genomic response to injury, which shifts the balance of outcome from cell death to cell survival.3
Microarray genome survey analysis suggests this genetic switch mimics the pattern seen in hibernation.3
Thus, as in hibernation, preconditioning elicits endogenous genetic adaptations that confer tolerance to the injurious effects of oxygen deprivation.
Of particular interest is the effect of delayed preconditioning on aspects of the inflammatory response,4
because inflammatory processes are increasingly being recognized as crucial propagators of ischemic brain injury. Rodent studies have recently demonstrated that although the total cellular infiltrate into the infarct zone is not different after preconditioning, the number of activated neutrophils is significantly reduced, as determined by reduced CD11b expression.4
The notion that delayed preconditioning might alter the profile of the systemic inflammatory response is of huge significance to CPB-related injury and clearly is an exciting area for future study.
The application to functional models represents another avenue for future investigation. We chose to use a model quantifying histologic neuronal injury because this remains a frequently used and reliable index of ischemic brain injury. Fluoro-Jade is especially sensitive for injured neurons, greatly facilitating their detection. Others have instead used markers of apoptosis or even functional outcome. Although difficult and crude in swine, functional models represent the proof of concept that we are intending to pursue in future studies. Microarray genome survey has additionally proved a valuable tool for both quantifying response to injury and providing insight into the putative mechanisms of preconditioning,3
and we are actively pursuing this technique in swine models at present.
One obstacle to the clinical application of LPS preconditioning is the risk posed by administering a potentially lethal substance to clinically vulnerable patients, irrespective of the tiny doses necessary. Interestingly, specific ligands for individual TLR subtypes have been shown to display cross-tolerance.30
For example, lipoteichoic acid preferentially binds TLR-2 and will induce tolerance to LPS (TLR-4) and visa versa. Future efforts should therefore explore the possibility that additional less noxious TLR ligands can be identified that confer similar protection to LPS through TLR cross-tolerance.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. P. Simon, R. Meller, A. Zhou, and D. Henshall Can Genes Modify Stroke Outcome and By What Mechanisms? Stroke, January 1, 2012; 43(1): 286 - 291. [Full Text] [PDF] |
||||
![]() |
S. L. Stevens, P. Y. Leung, K. B. Vartanian, B. Gopalan, T. Yang, R. P. Simon, and M. P. Stenzel-Poore Multiple Preconditioning Paradigms Converge on Interferon Regulatory Factor-Dependent Signaling to Promote Tolerance to Ischemic Brain Injury J. Neurosci., June 8, 2011; 31(23): 8456 - 8463. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Marsh, S. L. Stevens, A. E. B. Packard, B. Gopalan, B. Hunter, P. Y. Leung, C. A. Harrington, and M. P. Stenzel-Poore Systemic Lipopolysaccharide Protects the Brain from Ischemic Injury by Reprogramming the Response of the Brain to Stroke: A Critical Role for IRF3 J. Neurosci., August 5, 2009; 29(31): 9839 - 9849. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |