Improved regional cerebral blood flow is important for the protection seen in a mouse model of late phase ischemic preconditioning
Introduction
Ischemic stroke is a leading cause of morbidity and mortality. Therapy aims to alter the severity, susceptibility, and duration of the insult. The current clinical therapy of acute thrombolysis can alter the severity and the duration; however, the susceptibility of the brain remains unchanged. Targeting susceptibility is thus an important goal of experimental stroke research. Current efforts to alter the susceptibility of the brain are focused on ischemic preconditioning (IPC), which has been shown to improve neuronal survival in rodent brains (Kirino et al. 1991; Kitagawa et al. 1990 1991). IPC is a mechanism to induce tolerance to cerebral ischemia whereby a short, sub-lethal ischemia primes the brain to a more severe, injurious ischemia (Dirnagl et al., 2003). IPC has been observed in both global and focal ischemia, and has two time windows for protection; there is an early time window, occurring within minutes of the preconditioning ischemia/reperfusion, and a more delayed and prolonged time window, which is seen at 48–72 h post preconditioning ischemia/reperfusion and lasts 1 week (Perez-Pinzon et al., 1997). The protection observed during early and late phase preconditioning is proposed to have different mechanisms: the early phase preconditioning seems to protect the cells by altering the permeability of ion channels and post-translational modification of proteins, while the late phase preconditioning is dependent on protein synthesis (Gidday, 2006).#
Many proteins and pathways have been implicated as important for the tolerance induced by IPC, and also in the alteration of the susceptibility of the brain to ischemia. These include differential regulation of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor GluR1 and GluR2 subunits (Sommer and Kiessling, 2002), inhibition of anoxic long-term potentiation (LTP) (Kawai et al., 1998), and a reduced Ca2+ elevation in the CA1 pyramidal neurons(Shimazaki et al., 1998). As well, genomic analysis has been carried out to identify mechanisms for the protection. It was observed that transcripts important for metabolism were altered following ischemic preconditioning and that the profile was similar to that of an animal in hibernation (Stenzel-Poore et al., 2004).#
Investigations into transcript alterations as a means for understanding altered susceptibility of the brain are flawed if basic physiological parameters, such as regional cerebral blood flow (rCBF), are different between experimental groups. Previous studies have not documented an improved rCBF during injurious ischemia, due to preconditioning. LPS induced preconditioning (Dawson et al., 1999), distal MCAo IPC (Chen et al., 1996), cortical spreading depression (CSD) (Otori et al., 2003), and proximal MCAo IPC (Alkayed et al., 2002) were all shown to exert their protective effects independently of improved CBF. However, it was also shown that the microvasculature of the LPS preconditioned tolerant brain was better preserved and it was suggested that the secondary recovery in this model was due to maintenance of microvascular patency (Dawson et al., 1999). As well, recent reports suggest that regulation of rCBF is critical for the tolerance observed in the delayed form of IPC (Hoyte et al. 2004; Zhao and Nowak 2004 2006). The current study observed improvement in rCBF in the cohort of mice that had IPC, showing the importance of rCBF in IPC.#
Results
Ischemic preconditioning causes an improved regional cerebral blood flow
A significant reduction in blood flow was seen in naive (i.e. not preconditioned) animals upon introduction of the occluding filament into the origin of the MCA; this reduction in flow persisted for the entire occlusion period. In naive animals the flow was reduced to 16.3±3.6% (MCAO_45) and 17.1±8.6% (MCAO_15), while the preconditioned cohort rCBF was 33.9±13.2% of baseline during occlusion (p=0.001) (Fig. 1). The rCBF during occlusion of sham preconditioned (i.e. sham preconditioning ischemia, with injurious ischemia 72 h post) animals were not significantly different from that of naive mice (data not shown), indicating that this effect is not due to the isofluorane anesthetic and surgical stress.#
Arterial spin labeling and perfusion weighted MRI (MRP) demonstrate that ischemic preconditioning improved rCBF during injurious ischemia
MRP was used to evaluate the perfusion of the brain during occlusion in a striatal slice 2 mm thick through the area of the brain affected by the intraluminal suture model. Arterial spin labeling was used to tag the blood entering the brain and evaluate the cerebral perfusion. To facilitate comparison, the perfusion of the hemisphere ipsilateral to the MCAO was converted to a percentage of the contralateral hemisphere. The perfusion of the ipsilateral hemisphere of preconditioned mice during injurious occlusion was found to be 38.7±6.8% of the contralateral hemisphere, while the naive cohort was 26.7±5.8% of contralateral blood flow (p=0.04) (Fig. 2A). Representative MRP scans are shown in Fig. 2B. This set of experiments underscores the importance of ischemic preconditioning on rCBF in the intraluminal suture model of MCAO. These MRI experiments conclusively showed that IPC caused an improved perfusion during occlusion in preconditioned, as compared with naive mice.#
Infarct size in preconditioned mice is reduced as compared to naive mice
Percentage of brain infarcted was evaluated at 24 h. The infarcts were shown to be significantly smaller following IPC; MCAO_45 had 17.2±6.2% of the brain infarcted, while IPC_45 had only 5.1±4.6% of the brain infracted (p=0.003) (Fig. 3). Sham animals showed similarly sized infarcts as naive mice. These results confirmed that IPC protected the brain from injurious ischemia; similar protection was not induced by isofluorane anesthesia or surgical stress.#
Discussion
The two main findings from this study are: (1) IPC improves rCBF during injurious MCAO in the mouse, and (2) IPC results in significantly smaller infarcts following injurious ischemia in preconditioned, as compared with non-preconditioned, naive, mice. The LDF probe was applied to the parietal cortex, and this was the area that recorded the improved flow following IPC. This was also the region of the brain that was spared from infarction at 24 h survival. Previous studies had not shown improved CBF to be a mechanism for the protection afforded by IPC. The current study was likely able to visualize this difference due to the use of the murine model, and due to the placement of the LDF probe over the parietal cortex, the area of the brain spared by IPC. As well, arterial spin labeling MRP was used during ischemia to dynamically visualize the improved perfusion during injurious ischemia.#
Improved CBF as a mechanism for the protection of IPC is still a controversial topic in the field of ischemic preconditioning, as many previous studies documented a lack of improved rCBF during injurious ischemia in preconditioned animals; with various preconditioning stimuli, including: ischemic preconditioning (Alkayed et al. 2002; Chen et al. 1996), LPS (Dawson et al., 1999), and CSD (Otori et al., 2003). Currently, there are strong indications in the literature that improved rCBF could be an important aspect of the tolerance induced by IPC. The current study is supported by a recent report by Zhao and Nowak (2006), which highlighted the improved CBF due to IPC following permanent MCAO in the spontaneously hypertensive rat. This group used both laser Doppler flowmetry and iodo-antipyrine to demonstrate that CBF was improved at 3 h post occlusion; rats that had received IPC had significantly smaller infarcts as compared with naive following permanent MCAO.#
As well, previous studies have shown upregulated expression of proteins or pathways that have important vascular effects, following IPC. These include increased activity of eNOS and Akt, particularly within the vasculature (Hashiguchi et al., 2004), and specific localization of Akt to penumbral areas of the brain (Nakajima et al., 2004), and increased iNOS immunoreactivity in cerebral blood vessels at 24 h following a preconditioning insult (Cho et al., 2005). Activation of eNOS augments blood flow to the ischemic penumbra, and decreases leukocyte/endothelial and platelet/endothelial interactions (Atochin et al., 2003). IPC in both the early and late time windows has been shown to improve endothelial integrity, leading to less vascular stress, improved recovery of CBF, and lower levels of cerebral edema; this effect was inhibited by blockade of NOS (Vlasov et al., 2005). Knockout mice for either eNOS or nNOS showed an inability to demonstrate protection from injurious ischemia following IPC (Atochin et al., 2003), highlighting the necessity of the NOS system in the protection of IPC.#
IPC has various effects on thrombostasis, which could affect the perfusion of brain tissue during injurious ischemia. IPC has been shown to increase bleeding times of mice, and decrease platelet counts (Stenzel-Poore et al., 2003, 2004). As well, the vasculature of preconditioned brains is better preserved, with less BBB breakdown in the penumbral areas and reduced endothelial hsp70 expression (Masada et al., 2001).#
Measurements of rCBF must be carefully conducted, and LDF results must always be interpreted cautiously. Using Perimed equipment, LDF is a measure of relative cerebral blood flow and preconditioning has been shown to alter the baseline rCBF during future MCAo surgeries (Otori et al., 2003). The current study examined the cerebral perfusion with MRP to show that in the improved rCBF was not just an artifact of the LDF method; it was shown that mice that had received IPC had improved perfusion during the injurious ischemia, as compared with naive mice. It is clear that current studies into ischemic preconditioning, and specifically investigations into transcriptional differences induced by any intervention are fatally flawed if altered physiology is not taken into account.#
The improvement of rCBF due to treatment has been seen with various drug interventions (Table 1). Specifically, both competitive and non-competitive NMDA antagonists have been shown to improve blood flow in rats; this was seen in conscious rats (Roussel et al., 1992) and during MCAO (Buchan et al., 1992). As well, growth factors, such as GM-CSF, have been shown to exert their effects by improving the vascular networks, and specifically inducing arteriogenesis, leading to improved hemodynamics (Buschmann et al., 2003). It is clear that drug studies are flawed if they claim improvement in outcome solely on the action of the drug on its original target, if physiological changes are evident. Improved CBF can be an important and unconsidered effect of administering a drug. This will confound results if examinations of specific molecular targets independent of blood flow are being conducted. Altered physiology can be an important way to improve outcome in rodents following stroke, but new therapies must always examine the physiology with great care to ensure that the protection observed is translatable to humans. Changes in molecular pathways and alterations in transcript levels that suggest altered susceptibility of the brain to ischemia must be carefully studied in light of the improved rCBF that can occur following IPC. Neglecting the importance of physiological changes will render future studies into changing the susceptibility of the brain to ischemia faulty. This will weaken any attempts to translate these findings to human neuroprotection trials.#
A caveat of this study is that infarct size was examined at 24 h following injurious ischemia. Previous work has suggested that ischemic preconditioning simply postpones the maturation of injury in the gerbil model of global ischemia (Dowden and Corbett, 1999). It will be important to study the protection afforded by late phase preconditioning following long-term survival in the focal model.#
Conclusion
IPC induces tolerance to cerebral ischemia by alterations in multiple pathways and protein levels leading to a situation where there is less excitotocity, lowered inflammatory response, enhanced ability to deal with ROS, and improved recovery of LTP and protein synthesis. However, this study shows that, in the mouse, improved rCBF during the injurious ischemic insult is a key mediator of this protection. This study clearly reveals the importance of regional cerebral blood flow in the mouse model of ischemic preconditioning, and indicates the importance of stringent monitoring of rCBF in all future investigations of IPC.#
Experimental procedures
Animals
Three-month-old C57Bl/6 mice were obtained from Charles River (Ontario, Canada) within the weight class of 25–35 g. All procedures were approved by the animal care committee at the University of Calgary and followed the guidelines of the Canadian Council of Animal Care. Mice were stored in the single barrier unit at the University of Calgary, and were removed prior to preparation for surgery, where they remained until euthanized.#
Core temperature telemetry probe implantation
Sterilized core temperature telemetry probes (TA10TA-F20, Transoma Medical) were implanted into the peritoneal cavity seven days prior to middle cerebral artery (MCAO) under isofluorane anesthesia (induction 3% isofluorane in 30% O2 and air; maintenance 1.5% isofluorane in 30% O2 and air). During recovery, pain relief was afforded by subcutaneous (SC) injection of 0.05 mg/kg butorphanol every 4 h, as needed.#
Twenty-four hours prior to MCAO, mice were placed on receivers (RLA-1020, Data Sciences Int) and temperature was sampled every 30 s using the computerized temperature control system ART-2.2. This allowed for temperature and activity to be sampled in the freely moving mouse and for temperature to be regulated (Barber et al., 2004).#
Transient focal ischemia (MCAO)
MCAO was performed as described elsewhere (Barber et al. 2004; Hata et al. 1998; Longa et al. 1989). Briefly, isofluorane anesthesia was induced (3% initial, 1–1.5% maintenance) in 30% O2 and air. Under the operating microscope, the left common carotid artery (CCA), the left external carotid artery (ECA), and the left internal carotid artery (ICA) were isolated and a 6-0 suture was tied at the origin of the ECA and at the distal end of the ECA. The left CCA and ICA were temporarily occluded. The silicon-coated nylon suture was introduced into the ECA and pushed up the ICA until resistance was felt; the filament was inserted approximately 9–10 mm from the carotid bifurcation, effectively blocking the MCA. The diameter of the tip of coated suture was considered acceptable between 180 and 220 μm (Hata et al., 1998). The suture remained inserted for 15 min (preconditioning ischemia) or 45 min (injurious ischemia), after which it was removed and the ECA was permanently tied. During recovery, pain relief was afforded by SC injection of 0.05 mg/kg butorphanol every 4 h, as needed. CBF and infarct size were examined following injurious ischemia in naive animals (MCAO_45, n=10), and injurious ischemia in preconditioned animals (IPC_45, n=8). Laser Doppler flowmetry was also conducted during the preconditioning ischemia, and is displayed in Fig. 1 as MCAO_15. As well, CBF was examined with magnetic resonance imaging (MRI), animals were separated into two groups: naive (n=3), imaged during injurious occlusion and preconditioned (n=3), imaged during injurious occlusion.#
Ischemic preconditioning
In the case of preconditioning, the suture was inserted as described above for 15 min. The suture was then removed completely and the wound was sutured. At 72 h later, the animal was re-anesthetized and the MCAO surgery performed again, but this time a 45 min occlusion was induced.#
Sham-preconditioning with injurious ischemia
In the treatment group assigned to receive sham-preconditioning with injurious ischemia, the mice were anesthetized and head and neck area shaved as described above. The surgical procedure for MCAO proceeded as described above, up to and including the isolation of the left CCA, ECA, and ICA from the vagus nerve. There was no occlusion of the CCA and ICA and no opening made in the ECA. Instead, the wound was sutured closed and the mouse allowed to awaken from the anesthetic and given free access to food and water. Injurious MCAO was conducted at 72 h post recovery.#
Temperature regulation
All mice were regulated to maintain core body temperature of 36.5 °C during surgery, occlusion, and the immediate reperfusion period using a 125 W infrared heating lamp, as described previously (Barber et al., 2004). Following surgery, mice were regulated to maintain a core temperature of >34 °C by using a 125 W infrared heating lamp. For MRI studies, temperature regulation was conducted using rectal probes, due to the magnetic properties of the telemetry implants (Barber et al., 2005).#
Duration of survival and infarct analysis
Animals were euthanized by cervical dislocation 24 h following the injurious MCAO. The mice were decapitated and the brains removed under aseptic conditions. Brains were sectioned on a cryostat microtome at 20 μm and affixed to slides for hematoxylin and eosin staining. Distinct infarcted area was traced in serial brain sections through the striatum under light microscopy using Image Pro-Plus software. Percentage of brain infarcted was determined by using the following formula accounting for edema: % Brain infarcted=[Z−(X−Y)]/Y, where Z=area of infarct, X=area of ipsilateral hemisphere, and Y=area of contralateral hemisphere.#
Determination of rCBF
Measurement of CBF using laser Doppler flowmetry (LDF)
Transcranial measurements of CBF were made by LDF. A 0.5 mm diameter micro-fiber laser-Doppler probe (Probe 418, Perimed, Stocholm, Sweden) was attached to the skull with cyanoacrylate glue 6 mm lateral and 1 mm posterior of bregma. While under general anesthesia, rCBF was monitored within the parietal cortex. Blood flow velocity was measured pre-ischemia, during MCAO and reperfusion using data collection software (Perisoft Version 1.3 Perimed, Inc.). The rCBF during occlusion represents the average LDF reading normalized to the baseline reading for that animal; baseline readings are recorded at the start of surgery. The occlusion is thus displayed as a percentage of the baseline value.#
Measurement of CBF using magnetic resonance perfusion (MRP) imaging and arterial spin labeling
Animals were imaged as described previously (Qiao et al., 2004). Briefly, mice were imaged during injurious ischemia using a quadrature volume coil in a temperature controlled environment and an MR imaging system equipped with a 9.4T/21-cm horizontal bore magnet (Magnex) and an Avance console (Bruker, Germany). The head was restrained with ear pins, and respiration was monitored continuously. Images were acquired with a field of view of 2 cm2 and a data matrix of 256×128 for 7 slices 1 mm thick through the cerebrum. An ADC map and T1 maps were acquired along with perfusion imaging in the same slice using an arterial spin labeling technique (Qiao et al., 2004). CBF (in ml/g/s) was calculated as CBF=l(1/T1+d)(Mbcon−Mbinv)/2aMbcon, using “l”, the blood–brain barrier partition coefficient, as 0.9; “d” as 0.039s−1 and “a” as 0.75. Measured values included: T1 of brain; Mbcon, the intensity within control images; Mbinv, the intensity within images after arterial inversion of blood in the common carotid artery. MR parameters were measured over the ispilateral and contralateral hemispheres. The MR perfusion of the ischemic hemisphere was displayed as a percentage of the contralateral to facilitate comparison between groups.#
Statistics
For all experiments with parametric data and two treatment groups, Student's t-test was used. The Holm–Sidak method was used for pairwise multiple comparison procedures. Data are presented as mean±SD, with a value of P<0.05 considered to be statistically significant.#
Acknowledgments
LCH: Alberta Heritage Foundation for Medical Research Graduate Studentship and Canadian Institutes of Health Research Doctoral Studentship.#
MP: Wellcome Trust Research Fellowship.#
PAB: Canadian Institutes of Health Research, Alberta Heritage Foundation for Medical Research, and Heart and Stroke foundation of Canada.#
AMB: Canadian Institutes of Health Research, Heart and Stroke foundation of Canada, Canadian Stroke Network, Medical Research Council UK.#
Figures and Tables
Table 1
| Drug | Action | Effect on CBF | Reference |
| CGS-19755 | Competitive NMDA antagonist | Significant increases in several brain structures during MCAO (rat) | (Takizawa et al., 1991) |
| MK-801 | Non-competitive NMDA antagonist | 1) Significant increases in ischemic hemisphere during MCAO (rat) | 1) (Buchan et al., 1992) |
| 2) Large increases in conscious rats | 2) (Roussel et al., 1992) | ||
| 3) Dose-dependent vasodilation in cerebral microvasculature | 3) (Gelb et al., 1995) | ||
| GM-CSF | Growth factor | Triggered arteriogenesis, improved hemodynamic parameters | (Buschmann et al., 2003) |
| These drugs caused specific increases in rCBF to the ischemic hemisphere during MCAO. These included NMDA receptor antagonists, both competitive (CGS-19755) and non-competitive (MK-801), and growth factors (GM-CSF). |
References
13. J.M.GiddayCerebral preconditioning and ischaemic toleranceNat. Rev., Neurosci.72006437448
31. M.P.Stenzel-PooreS.L.StevensR.P.SimonGenomics of preconditioningStroke35200426832686