Stroke is a leading cause of death and disability in the United States, for which effectivetreatments remain elusive. Astrocyte survival is likely an important determinant of stroke volume andneuronal survival as loss of astrocyte viability would be predicted to augment stroke damage byglutamate release, collateral damage, and loss of their neuronal supportive functions. Thus, dissectingthe molecular mechanisms by which astrocyte viability is altered during stroke would be an importantadvance. A principal mechanism by which astrocytes adapt to the hypoxic environment during strokethrough the actions of hypoxia inducible factor 1 alpha (HIF-1a). HIF-1a, the master regulator of thecellular response to hypoxia, is a transcription factor stabilized and activated during hypoxia. HIF-1amarkedly increases the expression of multiple genes including those that would be predicted to improvesurvival and ones that would have a pro-apoptotic function. The conditions that control the pro-survivalor pro-apoptotic functions of HIF-1a remain undefined. Preliminary data suggests that HIF-1apro-apoptotic functions predominate in astrocytes during severe hypoxic stress enhancing cell death. Themolecular mechanisms by which HIF-1cc induces cell death is unknown but may involve induction ofpro-apoptotic members of the Bcl2 family Nip3, Nix, or Noxa. Furthermore, by associating with thepro-apoptotic transcription factor p53, HIF-1a may increase the pro-apoptotic functions of p53 duringsevere hypoxic stress. In contrast, pro-survival functions may be selectively induced during mildhypoxia. In fact, several reports suggest that HIF-1a plays a prominent role in the neuroprotectionafforded by the mild hypoxic stimulus of hypoxic preconditioning (HPC). Furthermore, hypoxia mimetic(HM) compounds, which induce HIF-1a protein and transactivation of HIF-1a target genes, areneuroprotective when used as a preconditioning stimulus and are actively being pursued as stroketherapeutics. This proposal will evaluate the role of HIF-1a in mediating pro-survival or pro-deathfunctions in astrocytes during stroke, following HPC, and with administration of HM compounds. Wepostulate that HIF-1a function in astrocytes contributes to astrocyte cell death during the severehypoxia and ischemia present during stroke. In contrast, we predict that during the mild hypoxic stressof hypoxic preconditioning, HIF-1a function in astrocytes mediates improved astrocyte and neuronalviability. Using in vitro and in vivo methodology, we will examine these postulates.PHS 398 (Rev. 09/04) Page 107 Form Page 2PO1 NS050315 Nedergaard, Maiken/Federoff, Howard Project 2Response to Critiques Project 2 (former project 3)Introduction We would like to thank the committee for the careful review of our application titled 'HIF-la function inastrocytes: deciphering pro-death and pro-survival roles during stroke.' Since the submission of the previousapplication, access to new transgenic mice have strengthened the proposed experimental paradigms, recentexperimental results have slightly altered the focus Aim II, and new experiments were added to Aims I & III,which strengthen the collaboration between our project and those proposed by Dr. Nedergaard. We willaddress these topics in order. New transgenic mice mouse models will enhance our experimental paradigms considerably. First, atransgenic mouse was provided to us through a collaboration with Dr. Frank Kirchhoff, who developed atransgenic mouse with astrocyte specific expression of a tamoxifen regulated ere recombinase (GFAP-CreERT2 mice). By breeding these mice with the HIF-laf+/f+, we will eliminate expression of HIF-la selectivelyin astrocytes after the mice reach adulthood, eliminating concerns of developmental abnormalities.Furthermore, we will evaluate the effect of astrocyte-specific loss of HIF-la function and its effect on targetstroke volume (Aim lib), and neuroprotection provided by hypoxic preconditioning (Aim Illb). These results canthen be compared to loss of HIF-la function selectively in neurons (HIF-la ::SynCre) or in neurons andastrocytes (HIF-laf+/f+::hGFAPcre) as previously proposed in the grant. Similarly to above, since the lastrevision of this project, we procured mice with universal expression of a tamoxifen-regulated ere recombinase[B6.Cg-Tg(cre/Esr1) 5Amc (Jax labs) [111, which will be referred to as the EsrCre mouse. By breeding thesemice with the HIF-laf+/f+ mice (HIF-la ::EsrCre), we eliminate HIF-la function in astrocyte cultures. Thisprovides us with astrocyte cultures with loss of HIF-la function (tamoxifen treated) that otherwise have theidentical genotype as controls (same cultures, not treated with tamoxifen). In the last revision of this grant, we provided preliminary data suggesting that astrocytes with loss ofHIF-la function are protected from oxygen glucose deprivation (OGD) compared to wildtype controls. While wereplicated this finding three times prior to grant submission, subsequent experiments demonstrated nodifference in OGD mediated astrocyte death with loss of HIF-la function. The original errant result was due tosubtle differences in density of astrocyte cultures with loss of HIF-la function compared to controls. Now,utilizing astrocyte cultures derived from HIF-laf+/f+::EsrCre mice, we induce selective loss of HIF-la function inastrocyte cultures derived from the same embryo and plated under identical conditions removing anyconfounds of subtle alterations in proliferation or other effects that may alter culture density betweengenotypes. Interestingly, using this approach, loss of HIF-la function protects astrocytes in culture from celldeath mediated by hypoxia/acidosis (see Preliminary Data). Examining hypoxia-acidosis mediated cell deathwas already proposed in our project, but now it will be the principal focus of Aim IIA, rather than the originalfocus on OGD. Finally, we added new experiments (Aim 1Cand AimlllB) to address a possible link between HIF-laand adenosine and their roles in hypoxic preconditioning. This provides a clear connection between our projectand Dr. Nedergaard's proposal, which in part examines the role of adenosine derived from astrocyte-releasedATP during stroke (Project 1).
In Aim 1 C,we explore the interesting possibility that hypoxic preconditioningalters gene expression through HIF-la and adenosine dependent molecular responses that interact to mediateneuronal protection. Recent evidence demonstrates that hypoxic stabilization of HIF-la decreases uptake ofextracellular adenosine[2], and enhances conversion of extracellular ATP to adenosine[3]. Both of these HIF-la dependent responses increase the concentration of extracellular adenosine and would be predicted toenhance adenosine mediated neuroprotective effects[3]. Thus, any putative HIF-la mediated neuroprotectioninduced by hypoxic preconditioning could be mediated in part through adenosine.
In Aim 1 MB, we will test if A1receptor antagonists alter hypoxic preconditioning utilizing the MCAO model.Responses to specific comments from the reviewers:From Reviewer: It is not clearly stated which targets will be evaluated by quantitative PCR. An interestingcomparison will be made between astrocyte cultures grown in 21% O2 compared to cultures grown in 5% O2.This is relevant because brain is normally subjected to oxygen tension in the 5% range, so the baseline level ofactivation of HIF-1a may be different in these conditions. In addition, experiments will be performed with andwithout cobalt or DMOG, two hypoxia mimetics. It is unclear how much of the effect of the HM is due tostabilization of HIF-1a vs. other effects. The suggestion is that pro-apoptotic targets are more dependent onHIF-1a than some of the pro-survival targets, and some data to this effect is shown in preliminary data. Astrength of the proposal is assessment of protein expression in addition to quantitation ofmRNA. 108
Showing the most recent 10 out of 39 publications
