Accumulating evidence indicated that sublethal pre-treatments can provide neuroprotective adaptation against subsequent severe ischemia in the brain (Kitagawa et al. 1990; Kapinya 2005; Ran et al. 2005). This process has been termed pre-conditioning or ischemic tolerance (IT). Evidence for the existence of ischemic pre-conditioning in humans has been reported (Moncayo et al. 2000; Schaller 2005). The major goal of studying IT is to identify the underlying endogenous protective signaling cascades, with the long-term goal to allow therapeutic augmentation of the endogenous protective mechanisms in cerebral ischemia and possibly to induce a protected state of the brain in conditions in which brain ischemia can be anticipated, for example, during surgery of the heart or cardiac arrest upon resuscitation.
To study the mechanisms of IT, clinically relevant models are needed. Experimentally, combinations of various types of ischemic insults have been developed to study IT (Kapinya 2005). Although ischemic pre-conditioning offers protection against ischemia in animal models, it could not be applied to patients for ethics reasons. On the other hand, moderate hypoxia, which does not cause neuronal death and may be safer to be applied in clinical practice, becomes an attractive method in animal research for IT. However, studies on hypoxia-induced IT mainly focused on focal ischemic models in the neonatal rats (Gidday et al. 1994; Ota et al. 1998) or adult mice (Bernaudin et al. 2002), it is not known whether hypoxia protects the adult rat brain against transient global cerebral ischemia (tGCI).
It has been suggested that hypoxic pre-conditioning (HPC) in the neonatal rat brain enhances intracellular pro-survival signaling pathways via altering gene expression and/or post-translational modification of signaling molecules (Ran et al. 2005). Akt (protein kinase B) is a protein kinase involved in survival signals as a downstream kinase of phosphoinositide 3-kinase (PI3K) in growth factor-mediated signaling cascades. Akt pathway is one of the pathways activated by ischemic pre-conditioning in the adult rodent brain, likely playing a critical role in promoting neuronal survival after ischemia (Yano et al. 2001; Yin et al. 2005). However, whether Akt pathway plays a role in hypoxia-induced tolerance to tGCI injury in the adult rat brain is not clear. Previous studies have indicated that activation of Akt promotes cell survival through phosphorylation of a series of substrates, such as class O members of the forkhead transcription factor family (FoxOs) (Brunet et al. 1999). FoxO is a mammalian homologue of DAF-16, which is known to regulate the life span of Caenorhabditis elegans and includes subfamilies of forkhead transcription regulators, including FKHR (FoxO1), FKHRL1 (FoxO3a), and AFX (FoxO4) (Lin et al. 1997). It has been documented that the activation of FoxOs is involved in the mechanisms of cell death induced by ischemia (Kawano et al. 2002; Shioda et al. 2007b). However, whether the inactivation of FoxOs is involved in the mechanisms of IT brought about by HPC remains unknown.
The present study aims to develop an animal model of hypoxia-induced tolerance to tGCI in adult rats and further investigate whether Akt is activated and its substrates, FoxOs, are inactivated in this kind of pre-conditioning. We present for the first time functional evidence that phosphorylation of Akt and FoxOs are important mediators of protection in HPC.
- Top of page
- Materials and methods
Induction of IT after hypoxia has already been reported in vitro and in vivo. However, in the latter case, this kind of tolerance had been described against global ischemia in the neonate or focal ischemia in the adult brain. To our knowledge, this is the first study to examine HPC of tGCI in adult rats. In addition, this study reported the role of Akt/FoxO signaling pathway in IT. The major findings of our study were as follows: (i) the maximum neuroprotection against tGCI was observed with 30 min of hypoxia and 1 day interval between hypoxia and ischemia, (ii) HPC induced a significant increase of phospho-Akt and phospho-FoxOs after ischemia except for a transient decrease, and (iii) inhibition of phospho-Akt and phospho-FoxOs reduced the protective effect of HPC, which provided evidence for a role of Akt/FoxO signaling pathway in IT.
Several types of stress can induce IT, such as brief global or focal ischemia, hyperthermia, and spreading depression (Kirino 2002; Blanco et al. 2006). These types of pre-conditioning can never been employed in clinical setting because of safety concerns. Our histopathologic analysis indicated that a short period of HPC was non-injurious. Up to 120 min hypoxia, there was no evidence of delayed cellular injury of the CA1 pyramidal cells that is highly vulnerable to global hypoxic ischemic injury.
The duration of hypoxia may influence the outcome of ischemia. Our results show that reduction of the HPC duration to 30 min is still enough to decrease the cell damage. The duration of hypoxia described in the literature is approximately 1–5 h (Bernaudin et al. 2002; Prass et al. 2003), which is longer than our shortest hypoxic duration (30 min). Our results show that 30 min of hypoxia exposition is enough to induce tolerance, and that this neuroprotection cannot be improved by increasing the duration of hypoxia. Indeed, in our model, pre-treated for 180 min hypoxia, half of the animals died, and the surviving animals showed cell damage in the CA1 sector after tGCI.
There are two temporally distinct types of IT afforded by sublethal pre-treatment – immediate and delayed tolerance. IT found in the brain is usually of the delayed type (Kirino 2002). In animal models, Kirino et al. (1991) found that the protection of IT once induced was believed to last for a few days and to wane gradually until it disappeared. Zhang et al. (2008) discovered that a repetitive focal ischemic pre-conditioning stimulus could be titrated to delayed tolerance when the interval between the pre-conditioning and test middle cerebral artery occlusion (MCAO) was 1, 2, 3, or 4 days, but that the maximal protection was seen with a 3-day interval. In clinical practice, if a stroke occurs within the same vascular territory during a limited time window (1–7 days) after an appropriate transient ischemic attack, the stroke may be less severe and the outcome may be better than otherwise (Schaller 2005). In our present study, an interval of 1–4 days could lead to IT, but 5 days was too long to induce tolerance, and the maximal protection was observed on day 1 after pre-conditioning. These results suggest that the time intervals between pre-treatment and the subsequent severe ischemia influence tolerance induction. It may be because of that the delayed tolerance requires a time window to synthesize the necessary proteins and activate genes (Barone et al. 1998).
As HPC is non-invasive, simple to perform and reproductive for IT, our model could be very useful for further study on the mechanisms of IT in the adult brain. Despite extensive research, the protective mechanisms of HPC are still not well understood. The survival or death of cells depends on the balance between pro-survival and pro-apoptosis signaling pathways. Under pathological conditions, such as ischemia, this balance may be destroyed, which leads to apoptosis. However, in IT, a new balance may be established.
Akt pathway is an important pro-survival signaling pathway. Changes in the phospho-Akt have been reported after brain ischemia (Osuka et al. 2004; Endo et al. 2006). We tested the possibility that the HPC would change the activation of Akt after ischemia. The present results demonstrated that phospho-Akt was transiently higher in the non-preconditioned rats than in the pre-conditioned ones (4 h after reperfusion). Ischemia/reperfusion has been reported to produce reactive oxygen species leading to severe oxidative stress (Morita-Fujimura et al. 2001; Kim et al. 2002). A previous report has shown that ischemic pre-conditioning results in an increase in the activity of superoxide dismutase, an anti-oxidant enzyme (Toyoda et al. 1997). These finding suggest that the oxidative stress is attenuated after ischemia in the pre-conditioned rats because of the preconditioning-induced increased activity of superoxide dismutase. As phospho-Akt in neurons increases after exposure to hydrogen peroxide in a dose-dependent manner (Crossthwaite et al. 2002), the present finding that phospho-Akt was transiently higher in the non-preconditioned rats than in the preconditioned ones may be attributed to the preconditioning-induced anti-oxidant effect. Although phospho-Akt was higher in the non-preconditioned rats than in the pre-conditioned ones early after ischemia, it rapidly decreased thereafter in the non-preconditioned ones. In contrast, such a rapid decrease in the activation level of Akt was prevented in pre-conditioned rats, and the activation level remained high in the pre-conditioned rats until 48 h after ischemia compared with non-preconditioned ones. These findings suggest that phospho-Akt induced after ischemia is inhibited in the pre-conditioned rats. Our result is consistent with reports from other studies using ischemic pre-conditioning (Yano et al. 2001; Yin et al. 2005). Our results also showed that phospho-Akt was increased 24 h after hypoxia, correlating with the development of IT. Interestingly, the decrease of phospho-Akt after tGCI with LY294002 pre-treatment was found. Accordingly, LY294002 treatment before HPC significantly prevented the neuroprotective action of pre-conditioning, with concomitant elimination of Akt activation. In addition, LY294002 treatment itself did not cause neuronal cell death. These results suggest that Akt activation after hypoxia and hypoxia-ischemia plays an important role in the establishment of IT. Unlike the rapid changes in phospho-Akt after brain ischemia, our results showed that total Akt protein level did not change following ischemia with or without HPC in CA1 subregion, which is consistent with almost all other studies (Yano et al. 2001; Yin et al. 2007). The results suggested that ischemia modulates neuronal death and HPC mediates IT through phosphorylation of Akt rather than through regulating its protein level.
Downstream targets for Akt that underlie neuroprotection have not been identified in the neurons. Akt prevents cell death through phosphorylating and inactivating downstream factors. The current study focused on phosphorylation of the FoxOs including FKHR, FKHRL1 and AFX after ischemia with or without HPC. It has been reported that phospho-FKHR decreases at 0, 30 and 60 min after reperfusion following 5 min of forebrain ischemia in gerbil models (Kawano et al. 2002). Additionally, Zhao et al. (2005) observed that phospho-FKHR decreased from 30 min to 48 h after focal ischemia in Sprague Dawley rats. Shioda et al. (2007b) reported that phospho-FKHR and phospho-FKHRL1 decreased from 2–6 h after 90 min of MCAO in mouse. These studies suggest that phospho-FoxOs such as FKHR and FKHRL1 decrease persistently after brain ischemia. Such results had a little distinction from ours. Our data demonstrated that phospho-FoxOs decreased after ischemia except for a transient increase. We believe that these different findings may be because of the difference in animal age and the model of ischemia. Ischemia/reperfusion causes inactivation of Akt, thereby decreasing phospho-FoxOs. The ultimate persistent dephosphorylation of FoxOs may contribute to the ischemic induced neuronal death.
The effect of HPC on phospho-FoxOs is first documented here. Our data showed increased phospho-FoxOs by increased Akt activity following tGCI likely mediated the neuroprotection induced by HPC. By applying another approach different from HPC, pre- and post-treatments with bis (1-oxy-2-pyridinethiolato) oxovanadium (IV), Shioda et al. (2007b) reported that phospho-FoxOs played an important role in protecting neurons after transient MCAO in mouse. Although using different treatments, it is possible that like bis (1-oxy-2-pyridinethiolato) oxovanadium (IV), HPC induced neuroprotection through inhibiting reduced Akt and FoxOs phosphorylation after brain ischemia. Further, the PI3K inhibitor, LY294002 prevented phosphorylation of Akt and FoxOs 24 h after ischemia/reperfusion with HPC, followed by increased neuronal damage, which meant that LY294002 blocked the protective effect of HPC. These results suggest that activation of Akt leads to the inactivation of FoxOs that may mediate IT after HPC. Although the precise mechanism underlying inactivation of FoxOs leading to IT after HPC is unclear, phospho-FoxOs likely induce cell survival by down-regulating apoptosis-inducing factors such as Bim and Fas-ligand. Bim is one of Bcl-2 members, and causes apoptosis in various cell types. Bim is also expressed in neuronal cells (O’Reilly et al. 2000). Fas is a member of the tumor necrosis factor receptor family, and its ligand, Fas-ligand, plays important roles in apoptosis (Nagata 1997). Activation of Fas leads to formation of a death-inducing signaling complex composed of the Fas-associated death domain and pro-caspase 8. Pro-caspase 8 is proteolytically cleaved and consequently activates caspase pathways and thereby cells are led to apoptosis. Recent studies reported that Bim (Fukunaga et al. 2005) and Fas-ligand (Fukunaga et al. 2005; Shioda et al. 2007a,b) expression significantly increased after brain ischemia/reperfusion accompanyed by dephosphorylation of FKHR. Therefore, it is possible that the phosphorylation of FoxOs induced by HPC via activation of Akt can block the expression of Bim and Fas-ligand, leading to cell survival. Further experiments will be required to investigate this interesting possibility.
In summary, we present a strong evidence of neuroprotection against brain injury by tGCI in adult rats using a short period of hypoxia. Activation of the Akt/FoxO signaling pathway is associated with the HPC in adult rats by inhibition of dephosphorylation of phospho-Akt and phospho-FoxOs. To understand the mechanisms induced by hypoxia that lead to neuroprotection against ischemia would be very important for identifying potential novel therapeutic targets in the field of ischemic stroke.