Neuron-specific inactivation of the hypoxia inducible factor 1 alpha increases brain injury in a mouse model of transient focal cerebral ischemia

Abstract

In the present study, we show a biphasic activation of hypoxia inducible factor 1alpha (HIF-1) after stroke that lasts for up to 10 d, suggesting the involvement of the HIF pathway in several aspects of the pathophysiology of cerebral ischemia. We provide evidence that HIF-1-mediated responses have an overall beneficial role in the ischemic brain as indicated by increased tissue damage and reduced survival rate of mice with neuron-specific knockdown of HIF-1alpha that were subjected to transient focal cerebral ischemia. In addition, we demonstrated that drugs known to activate HIF-1 in cultured cells as well as in vivo had neuroprotective properties in this model of cerebral ischemia. This protective effect was significantly attenuated but not completely abolished in neuron-specific HIF-1alpha-deficient mice, suggesting that alternative mechanisms of neuroprotection are also implicated. Last, our study showed that hypoxia-induced tolerance to ischemia was preserved in neuron-specific HIF-1alpha-deficient mice, indicating that the neuroprotective effects of hypoxic preconditioning do not depend on neuronal HIF-1 activation.

Figure 1.

HIF-1α induction in the mouse brain after transient ischemia. A, Double immunostaining of HIF-1α (green) and NeuN (red) at 6 h of recovery from MCAo. Boxed area in cresyl violet-stained section indicates the area in which representative photomicrographs were taken. Top panels show images from cortex and striatum of non-ischemic and ischemic hemispheres. Scale bars, 20 μm. Bottom panels show higher-magnification views of neurons expressing HIF-1α that are located in the infarct area (a–c) and surrounding tissue (d–f). Scale bar, 50 μm. Merged images are shown in yellow. B, Western blot analysis of HIF-1α in tissue lysates from non-ischemic and ischemic hemisphere of mice subjected to MCAo followed by recovery (1 h to 8 d). Codetection of α-tubulin was performed to assess equal loading. Nuclear extracts from normoxic and hypoxic cells were used as positive (+) and negative (−) controls, respectively. Protein bands were quantified, and the value obtained from sham-operated control (S) was arbitrarily defined as 1. Graph shows data acquired from four independent experiments. *p ≤ 0.05 compared with sham.

Figure 2.

Neuron-specific knockdown of HIF-1α. A, Representative microscopic views of boxed area in diagram are shown. CaMKIIα–Cre recombinase transgenic line (R1ag#5) were mated with a LacZ reporter mouse (ROSA26Sor), and brains from offspring (P20) were analyzed by β-galactosidase staining (red). Double staining with NeuN (green) demonstrated neuronal β-galactosidase production. Bottom panels show higher-magnification views of the striatum and cortex. B, Southern blot analysis was performed on DNA isolated from the cerebral cortex, striatum, and cerebellum of Cre(+):HIF-1α^F/F mice at indicated ages (E16 to P90), to detect floxed undeleted allele (HIF-1α^F) and deleted allele (HIF-1α^Δ). C, Decreased cerebral HIF-1α expression at normoxic (C) and hypoxic (H) (8% O₂ for 4 h) conditions was confirmed in R1-HIF-1α^Δ/Δ mice by Western blot analysis. Heterozygous HIF-1α^Δ/wt also showed a moderate reduction of hypoxic HIF-1α levels. Data are expressed as the mean ± SD from four different experiments. *p ≤ 0.05 versus hypoxic wild-type (Wt-H). ^# p ≤ 0.05 versus hypoxic heterozygotes HIF-1α^Δ/wt.

Figure 3.

Cerebral vasculature analysis. A, Ventral view of large cerebral blood vessels of control (HIF-1α^F/F) and neuronal HIF-1α-deficient (R1- and L7-HIF-1α^Δ/Δ) mice that were perfused with India ink. B, Representative microscopic views of boxed area in diagram are shown. Brain sections from HIF-1α^F/F and R1- and L7-HIF-1α^Δ/Δ mice were immunostained for the endothelial cell-specific marker CD31 (green) for quantification of capillary numbers per area. *p < 0.05 compared with HIF-1α^F/F; n = 4.

Figure 4.

Role of neuronal HIF-1α in ischemic brain injury. A, Immunoblot analysis of HIF-1α in the ischemic hemisphere of R1-HIF-1α^Δ/Δ and control (HIF-1α^F/F) mice subjected to transient MCAo and recovery (1 h to 14 d). Codetection of α-tubulin was performed to assess equal loading. Protein bands were quantified, and the value obtained from sham-operated control (S) was arbitrarily defined as 1; n = 5. *p < 0.05 versus sham HIF-1α^F/F; ^# p < 0.05 versus sham HIF-1α^Δ/Δ. B, Evolution of the infarct volume was assessed at different time points of recovery after MCAo in the following genotypes: HIF-1α^F/F and R1- and L7-HIF-1α^Δ/Δ. At 4 d of recovery, infarct area was measured in 12 sequential sections taken at rostral to caudal regular intervals. Representative sections stained with cresyl violet are shown from the indicated genotypes at 4 d of recovery. C, Blinded scoring of neurological deficit was assessed at different time points of recovery in HIF-1α^F/F and R1- and L7-HIF-1α^Δ/Δ. Kaplan–Meier survival analysis after transient MCAo in the indicated mouse genotypes. In B and C, at least 23 mice per genotype are represented in the neurological score and survival curves. *p < 0.05 compared with HIF-1α^F/F.

Figure 5.

Cerebral blood flow and rectal temperature after MCAo. Local CBF was monitored before, during MCAo, and after reperfusion by laser Doppler flowmetry. Graphs indicate CBF monitored at the ischemic area (A) and the periphery (B). Pre-ischemic values were arbitrarily defined as 100%. Rectal temperature (degrees Celsius) was measured using a rectal probe at 12 h intervals for up to 4 d after ischemia (C). Data are expressed as mean ± SD from 8–10 animals per genotype.

Figure 6.

HIF-1-dependent expression of proapoptotic mediators after brain ischemia. A, Western blot analysis of the indicated proteins in the ischemic hemisphere of HIF-1α^F/F and R1-HIF-1α^Δ/Δ mice. Animals were subjected to transient MCAo and killed at different periods of recovery (1 h to 8 d). B, Real-time RT-PCR analysis of the indicated genes in the ischemic hemisphere of HIF-1α^F/F (filled circles) and R1-HIF-1α^Δ/Δ (open circles) mice. Data were normalized to β-actin and expressed relative to HIF-1α^F/F sham control (S), which was arbitrarily defined as 1. Data are expressed as mean ± SD from five animals per genotype. *p < 0.05 versus sham.

Figure 7.

Expression of HIF-regulated genes in the ischemic brain. Real-time RT-PCR analysis of the indicated genes in the ischemic hemispheres of HIF-1α^F/F (filled circles) and R1-HIF-1α^Δ/Δ (open circles) mice. Animals were subjected to transient focal ischemia and killed at different times of recovery (1 h to 8 d). Data were normalized to β-actin and expressed relative to HIF-1α^F/F sham control (S), which was arbitrarily defined as 1. Data are expressed as mean ± SD from five animals per genotype.

Figure 8.

Effect of prolyl hydroxylases inhibitors on ischemic brain damage. A, Western blot analysis of HIF-1α in brain cortex of animals treated with vehicle intraperitoneally (V), DP (20 mg/kg, i.p.), DHB (200 mg/kg, i.p.), and DFO (300 mg/kg, s.c.). Animals were killed 6 h after drug administration. Normoxic (N) and hypoxic (H) (8% O₂ for 4 h) animals were used for comparison. B, RT-PCR analysis of Vegf and Glut-1 in brain cortex at 24 h after treatment of animals (HIF-1α^F/F and R1-HIF-1α^Δ/Δ) as indicated in A. C, Pretreatment of HIF-1α^F/F and R1-HIF-1α^Δ/Δ mice with the indicated drugs was performed 6 h before the ischemic insult, and posttreatment began 6 h after MCAo. Evaluation of infarct volume and neurological deficit was made 4 d after the onset of ischemia. Data are presented as mean ± SD (n = 6). *p ≤ 0.05 compared with normoxia (N) or vehicle (V).

Figure 9.

Influence of hypoxic preconditioning on ischemic brain injury. Wild-type C57BL/6J (Wt), HIF-1α^F/F, and R1- and L7-HIF1α^Δ/Δ were exposed to normoxia (N; 21% O₂) or hypoxia (H; 8% O₂) for the indicated durations, 24 h before MCAo. Infarct volumes were analyzed at 4 d of recovery. *p < 0.05 between hypoxic and normoxic treatments.

Figure 10.

Increased HIF-2α and Epo expression in HIF-1α^Δ/Δ mice. A, Western blot analysis of HIF-2α in brain samples of animals subjected to normoxia (C) (4 h) or hypoxia (H) (8% O₂, 4 h). Protein bands were quantified and normalized to α-tubulin, and the value obtained from normoxic wild-type samples (C, Wt) was arbitrarily defined as 1. B, RT-PCR and Western blot analysis of Epo in animals subjected to the same experimental conditions as described in A. Data are expressed as mean ± SD from five animals for each genotype including wild-type (Wt), HIF-1α^F/F, and HIF-1α^Δ/Δ. *p ≤ 0.05 compared with normoxic wild-type (C, Wt). ^# p < 0.05 versus hypoxic wild type (H, Wt).