The Akt-Endothelial Nitric Oxide Synthase Pathway in Lipopolysaccharide Preconditioning-Induced Hypoxic-Ischemic Tolerance in the Neonatal Rat Brain
Background and Purpose— Low-dose lipopolysaccharide (LPS) preconditioning provides neonatal rats long-term neuroprotection against hypoxic ischemia (HI). Upregulating endothelial nitric oxide synthase (eNOS) protects against cerebral ischemia; however, whether eNOS is required for LPS preconditioning-induced protection in neonatal rats is unknown. We hypothesized that Akt activation, which upregulates eNOS in neurons and endothelial cells, is required for LPS preconditioning-induced tolerance against HI in the neonatal brain.
Methods— Six-day-old rat pups were intraperitoneally injected with LPS (0.05 mg/kg) or normal saline 24 hours before HI. Immunoblotting and immunohistochemistry were used to determine the phospho-Akt (pAkt Ser473), phospho-eNOS (peNOS Ser1177), and eNOS levels and immunofluorescence to determine the cellular distribution of eNOS and pAkt Ser473. Pharmacological and genetic approaches were used to regulate Akt and eNOS, and the weight loss of cerebral hemispheres on postnatal Day 21 was used to assess outcomes.
Results— eNOS, peNOS (Ser1177), and pAkt (Ser473) levels were significantly higher in LPS- than in normal saline-treated rats 24 hours postinjection. LPS-induced eNOS was expressed primarily in neurons and vascular endothelial cells. N-omega(ω)-nitro-l-arginine and antisense oligodeoxynucleotide treatment significantly reduced eNOS expression in neurons and endothelial cells and inhibited LPS-induced protection against HI in rat pups. l-arginine and adenovirus eNOS transfection upregulated eNOS and protected the rat pups against HI. Wortmannin treatment before LPS preconditioning significantly reduced eNOS expression in neurons and endothelial cells, which inhibited LPS-induced protection against HI.
Conclusions— Akt-mediated eNOS upregulation in neurons and vascular endothelial cells is required for LPS-induced tolerance against HI in the neonatal rat brain.
Hypoxic ischemia (HI) is a major cause of neonatal mortality and subsequent neurodevelopmental morbidity in surviving infants. A new approach to studying treatment for HI brain injury is investigating the state of tolerance.1 Tolerance is attained by preconditioning tissue to sublethal stress, which augments endogenous defense mechanisms against subsequent lethal insults. Clarifying the mechanisms of preconditioning may lead to new pharmacological therapies against HI in neonates.1,2
Endothelial nitric oxide synthase (eNOS) is found primarily in the endothelium of cerebral blood vessels. Nitric oxide generated by eNOS is crucial for vascular function and homeostasis. Using l-arginine, statins, or physical activity to upregulate eNOS increases cerebral blood flow and protects against cerebral ischemia.3 Lipopolysaccharide (LPS) preconditioning provides tolerance against ischemia in adult and neonatal rats.4–6 Systemic LPS injections upregulated eNOS rather than inducible NOS (iNOS) in astrocytes and blood vessels.7,8 However, whether eNOS is required for LPS preconditioning-induced HI tolerance in neonatal brain remains unclear. If eNOS is required, then the preconditioning-activated upstream pathway that leads to eNOS upregulation warrants investigation because the pathway will be an attractive therapeutic target for HI.
Akt, a pivotal molecule for cell survival and vascular homeostasis, is a key mediator of eNOS activation because it phosphorylates eNOS on Ser1177.3 Hypoxic preconditioning suppressed neuroinflammation in neonatal HI through Akt activation.9 Akt activation accounted for eNOS upregulation in vascular endothelial cells10 and in ischemic tolerance of gerbil hippocampus CA1.11 Our previous study showed that low-dose LPS (0.05 mg/kg) preconditioning provided long-term neuroprotection in neonatal rats.6 Whether Akt-eNOS signaling is critical in LPS preconditioning-induced HI tolerance in the neonatal brain remains undetermined. A shared signaling pathway may underlie preconditioning-induced protection in neurons and vascular endothelial cells.12 Therefore, we hypothesized that Akt-mediated eNOS upregulation in neurons and endothelial cells is required for LPS-induced HI tolerance in the neonatal brain.
Materials and Methods
Low-Dose LPS Preconditioning
This study was approved by the Animal Care Committee of National Cheng Kung University. Postnatal (P) Day 6 Sprague-Dawley female rat pups (purchased from BioLASCO Taiwan Co, Ltd) were intraperitoneally injected with LPS (0.05 mg/kg; Escherichia coli 0111:B4; Sigma-Aldrich) or pyrogen-free normal saline (NS) 24 hours before HI on P7.6 On P7, the rats were anesthetized (2.5% halothane), and then the right common carotid artery was permanently ligated. After they had been allowed to recover for 1 hour, the rats were placed in airtight 500-mL containers with humidified 8% oxygen flow rate of 3 L/min for 2 hours. The pups were sorted into 4 groups: NS (NS without HI), LPS (LPS without HI), NS-HI (NS 24 hours before HI), and LPS-HI (LPS 24 hours before HI).
On P21, the rat pups’ brains were sectioned at the midline, and the left and right hemispheres were weighed. The percentage of hemispheric weight loss, measured as [(left hemisphere weight−right hemisphere weight)/left hemisphere weight], was used as the measure of cerebral HI injury.6,13
Immunohistochemistry and Analysis
Immunohistochemistry for eNOS and pAkt (Ser473) was done 24 hours after LPS or NS injection. The brains were coronally sectioned (14-μm thick). The primary antibodies used were anti-eNOS (1:100; BD Transduction) and antipAkt (Ser473; 1:200; Cell Signaling). Biotinylated antirabbit and antimouse antibodies were the secondary antibodies (all 1:200). eNOS and pAkt expression was analyzed using integrated optical density in the cross-sectional areas of the cortex in 2 reference planes corresponding to plates 31 and 39 in a rat brain atlas.14 The analysis used imaging software (ImagePro Plus 6.0; Media Cybernetics) at 400× magnification per visual field (one visual field=0.096 mm2). Three visual fields in the cortex of the hemisphere per section and 2 sections per brain were analyzed and averaged.6
The sections were incubated overnight with a mixture of 2 of the following primary antibodies: anti-eNOS (1:1000), anti-pAkt (1:1000), antineuronal nuclear antigen (NeuN) (neurons; 1:100; Chemicon), antiendothelial cell antigen (RECA) (endothelial cells; 1:100; Abcam), antiglial fibrillary acidic protein (GFAP) (1:100; Chemicon), or terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling (TdT-fluorescein-FragEL DNA fragmentation detection kit; Calbiochem). The sections were then washed and incubated with fluorescein isothiocyanate-conjugated antimouse IgG antibody and Texas Red-conjugated antirabbit IgG antibody (1:500; Invitrogen). The fluorescence signals were detected and the results recorded (Eclipse E400 microscope; Nikon) at excitation-emission wavelengths of 596 to 615 nm (Texas Red, red) and 470 to 505 nm (fluorescein isothiocyanate, green).12
The cortex was homogenized, resolved using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and blotted electrophoretically to polyvinylidene fluoride membranes. The membranes were incubated with primary antibodies: anti-eNOS (1:1000), antiphospho-eNOS (peNOS, Ser1177; 1:1000; Cell Signaling), antineuronal NOS (nNOS; 1:1000; BD Transduction), anti-iNOS (1:800; Santa Cruz), anti-Akt (1:1000; Cell Signaling), antipAkt (Ser473; 1:1000), or antiactin (1:3000; Chemicon). Immunoreactivity was detected using horseradish peroxidase-conjugated IgG antibody (1:10 000; Calbiochem) and visualized using enhanced chemiluminescence (Millipore). VisionWorks LS (Ultra-Violet Products) analysis software was used for densitometry.
We first pretreated the rat pups, 30 minutes before LPS injection, with a nonselective constitutive NOS inhibitor N(ω)-nitro-l-arginine (2 mg/kg), the nNOS inhibitor 7-nitroindazole (7-NI; 40 mg/kg), or the iNOS inhibitor aminoguanidine (AG; 400 mg/kg; all from Sigma-Aldrich) intraperitoneally in doses sufficient to inhibit the respective NOS isoform as reported in a hypoxic preconditioning model of neonatal rats.15 All drugs except 7-NI were dissolved in NS; 7-NI was dissolved in 100% corn oil and then sonicated. We then administrated 7-NI (40 mg/kg) or AG (400 mg/kg) 30 minutes before and 12 hours after LPS or vehicle injection in rat pups.
P6 pups were intracerebroventricularly infused with eNOS antisense or scrambled oligodeoxynucleotides (ODN) in the right cerebral hemisphere using a 30-gauge needle on a 10-μL Hamilton syringe.12,13 The injection location was 2.0 mm posterior to and 1.5 mm lateral to the bregma and 2.0 mm beneath the skull surface. The infusion rate was 1 μL/min. The first ODN (2 μL, 25 μg) was injected 30 minutes before LPS and the second (2 μL, 25 μg) 12 hours after LPS. The following ODNs were used: antisense eNOS: 5′-C*TC TTC AAG TTG CCC AT*G-3′ and scrambled: 5′-GTC TTG AAC TTC CCG AT*C-3′ (*phosphorothioate linkage). Based on the cDNA sequence for rat eNOS (Genebank accession number NM_021838), the antisense sequence matched with rat eNOS cDNA sequence, but the scrambled ODN showed no significant matches. The cortices were collected 12 hours after the second ODN.
Twenty-four and 16 hours before HI, the pups were intraperitoneally injected with l-arginine (200 mg/kg), an eNOS agonist, or d-arginine (200 mg/kg; Sigma-Aldrich).3 The cortices for immunoblotting and the brain sections for immunohistochemistry were collected 16 hours after the second arginine injection.
Adenoviral eNOS Gene Transfer
The replication-deficient recombinant adenoviruses nuclear-targeted β-galactosidase, and eNOS were kindly provided by Professor S.K. Shyue (Institute of Biomedical Sciences, Academia Sinica, Taiwan). The constructed adenoviruses were replicated in 293 cells and harvested using CsCl gradient ultracentrifugation.16 The viral particles collected were dialyzed and viral titers determined using a plaque-assay method. These adenoviruses (1×109 plaque-forming units per injection) were infused into the right hemisphere. The pups were given 2 consecutive injections; the first injection was 2.0 mm posterior to and 1.5 mm lateral to the bregma and 2.0 mm beneath the skull surface, and the second injection was 2.0 mm posterior to and 2.0 mm lateral to the first injection site and 2.0 mm beneath the skull surface. These transfers occurred on P2, P4, and P6, respectively, and the cortices were collected on P7.
Phosphotidylinositol 3-Kinase/Akt Inhibition
A specific phosphotidylinositol 3-kinase (PI3K) inhibitor, Wortmannin (2 mmol/L in 2% dimethyl sulfoxide; Sigma-Aldrich) or 2% dimethyl sulfoxide was injected into the right cerebral hemisphere 30 minutes before and 16 hours after LPS and NS injection.11 The cortices for immunoblotting and brain sections for immunohistochemistry were collected 8 hours after the last injection.
Semiquantitative Reverse Transcription Polymerase Chain Reaction
Total RNA was extracted from the cortex using a reagent (Trizol; Invitrogen). Five micrograms of total RNA and 1.5 μg of oligo-dT primer were mixed and preincubated at 70°C for 10 minutes. This mixture was mixed with reverse transcription (RT) reagent containing 25 U of MMLV-reverse transcriptase, 10 μL of 5× reaction buffer, and 0.5 mmol/L of dNTP (Promega). Each polymerase chain reaction mixture contained 5 μL of RT product, 1 U of Taq DNA polymerase (Viogene), 2 μL of 10× polymerase chain reaction buffer plus MgCl2, 0.2 mmol/L of dNTP, 0.5 μmol/L of eNOS specific primers (forward primer sequence: 5′GAGAATTCCACCTCACTGTAGCTGTGCTGGCA3′ and reverse primer sequence: 5′TCGATTTCCCAGGGCACTGCGCCCCGCAACTG3′; product sizes: 395 bp), and actin (forward primer sequence: 5′ ACCCAGATCATGTTTGAGAC 3′ and reverse primer sequence: 5′TCTCTTGCTCGAAGTCTAGG3′; product sizes: 220 bp). The amplified reaction was done using a thermocycler for a single 3-minute initial denaturation at 94°C and then 28 cycles for eNOS or 22 cycles for actin under the polymerase chain reaction conditions.
Statistical significance (P<0.05) was determined using 1-way analysis of variance. Tukey significant difference method was used for post hoc comparisons in 1-way analysis of variance. Continuous data were means±SEM unless indicated otherwise.
Low-Dose LPS Upregulated eNOS
The LPS-HI group had a lower mortality rate (3% versus 16%) during HI in P7 and significantly less brain injury in P21 (P<0.001, Figure 1A) than the NS-HI group. Immunofluorescence at 24 hours after HI showed many terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling (TUNEL)(+) cells in the ipsilateral cortex of the NS-HI group but very few of the LPS-HI group. Most TUNEL(+) cells in the NS-HI group were neurons that colocalized with neurons, and endothelial cells that coexpressed endothelial cells (Figure 1B) but not astrocytes that expressed glial fibrillary acidic protein (Supplemental Figure I; available at http://stroke.ahajournals.org). This finding suggests LPS preconditioning protects neurons and vascular endothelial cells against HI injury. Immunoblotting showed that low-dose LPS significantly increased eNOS, but not nNOS or iNOS, in the cerebral cortex 12 and 24 hours postinjection (P<0.01, Figure 1C). Semiquantitative reverse transcription-polymerase chain reaction analysis revealed that LPS significantly upregulated eNOS mRNA at 12 (P<0.01) and 24 hours (P<0.05) postinjection (Figure 1D). The representative pictures showed the locations of cortical areas used for the quantitative measurement of eNOS and pAkt immunoreactivities (Figure 1E). Immunohistochemistry revealed weak eNOS immunoreactivity in the NS group but significantly higher eNOS in the LPS group (P<0.01). The LPS-induced eNOS expression was located primarily in neurons and vascular cells (Figure 1F). Immunofluorescence confirmed that most eNOS(+) cells in LPS-preconditioned pups were neurons and vascular endothelial cells but not astrocytes (Figure 1G).
eNOS Was Required for LPS Preconditioning-Induced Tolerance
LPS preconditioning-induced protection was significantly lower in pups pretreated with N(ω)-nitro-l-arginine (P<0.01), but not in pups pretreated with AG or 7-NI before LPS (Figure 2A). In contrast to the irreversible and long-acting property of N(ω)-nitro-l-arginine, both 7-NI and AG have short-acting properties. To further delineate the effect of nNOS and iNOS on both the early and delayed phase of LPS preconditioning, we next administrated 7-NI (40 mg/kg) or AG (400 mg/kg) 30 minutes before and 12 hours after LPS injection in rat pups. We found that LPS preconditioning-induced protection was still not lower in pups treated with AG or 7-NI before and after LPS (Figure 2B). Immunoblotting showed that eNOS antisense ODN significantly reduced eNOS in the ipsilateral cortex more than in the contralateral cortex 12 hours after the second ODN injection (P<0.001). Scrambled ODN had no significant effect on eNOS expression between the 2 hemispheres (Figure 2C). Immunohistochemistry showed that eNOS antisense markedly reduced eNOS expression (P<0.001) in neurons and vascular cells in LPS-preconditioned pups (Figure 2D). LPS-HI pups pretreated with eNOS antisense had significantly more brain damage than those with scrambled ODN (P<0.05; Figure 2E). LPS-HI pups pretreated with eNOS antisense had similar level of brain damage as NS-HI pups.
eNOS Upregulation by LPS Protected Against HI
Immunoblotting showed that eNOS expression was significantly higher in l-arginine- than in d-arginine- and NS-treated pups 24 hours postinjection (both P<0.01; Figure 3A). Immunohistochemistry revealed that eNOS expression was significantly higher in l-arginine- than d-arginine- and NS-treated pups (both P<0.01) and l-arginine-induced eNOS was localized in neurons and vascular cells (Figure 3B). l-arginine- (P<0.01) but not d-arginine-treated pups had significantly less brain injury compared with NS-treated pups (Figure 3C).
Immunoblotting showed that eNOS levels were significantly higher in the cerebral cortex 3 and 5 days after adenoviruses nuclear-targeted eNOS than after adenoviruses nuclear-targeted β-galactosidase (all P<0.05) or NS treatment (5 days, P<0.01; 3 days, P<0.05; Figure 4A). Pups pretreated with adenoviruses nuclear-targeted eNOS 5 days before HI had significantly less brain injury than pups pretreated with adenoviruses nuclear-targeted β-galactosidase or NS (all P<0.001; Figure 4B).
LPS Preconditioning Activated Akt
eNOS levels were significantly higher 12 (P<0.05) and 24 hours (P<0.01) after LPS injection, and peNOS (Ser1177) levels were significantly higher at 1 and 24 hours (both P<0.01) postinjection (Figure 5A). eNOS and peNOS levels were significantly higher 24 hours after LPS preconditioning. Compared with NS, LPS significantly upregulated pAkt (Ser473) at 0.5 (P<0.05), 1 (P<0.01), 3 (P<0.01), and 24 hours (P<0.05) postinjection. Immunohistochemistry confirmed that LPS preconditioning significantly induced pAkt (P<0.01) and that LPS-induced pAkt was found primarily in neuronal and vascular cells (Figure 5B). Immunofluorescence revealed that most LPS-induced pAkt(+) cells were neurons and endothelial cells (Figure 5C). In addition, the pAkt(+) neurons and endothelial cells coexpressed eNOS.
Akt-eNOS Activation Was Required for LPS-Induced Tolerance
The intracerebroventricular infusion of Wortmannin 30 minutes before and 16 hours after LPS significantly reduced pAkt compared with the infusion of dimethyl sulfoxide (P<0.01; Figure 6A). Wortmannin also significantly downregulated peNOS (Ser1177; P<0.01) and eNOS (P<0.05) levels. In LPS-preconditioned pups, Wortmannin significantly downregulated eNOS in neurons and endothelial cells (P<0.01; Figure 6B). In rat pups without preconditioning, brain injury was not significantly different in rat pups pretreated with Wortmannin and dimethyl sulfoxide 24 hours before HI. In contrast, in rat pups with LPS preconditioning, brain injury was significantly higher in pups pretreated with Wortmannin (P<0.05) than in pups pretreated with dimethyl sulfoxide (Figure 6C).
The present study provides evidence that low-dose LPS preconditioning in rat pups confers neuroprotection and vasculoprotection against HI injury by activating both Akt and eNOS in neurons and endothelial cells. We found that low-dose LPS increased eNOS, but not iNOS or nNOS, expression. The LPS-induced eNOS expression was localized primarily in neurons and endothelial cells. Inhibiting eNOS expression by N(ω)-nitro-l-arginine or antisense ODN showed that eNOS was required for LPS-preconditioning-induced HI tolerance. We also found that Akt activation was necessary to upregulate eNOS in neurons and endothelial cells as well as to provide protection against HI in LPS preconditioning. Furthermore, we showed that pharmacologically and genetically upregulating eNOS protected rat pups against HI. Taken together, these findings show that upregulating eNOS by activating Akt in the neonatal brain is critical for low-dose LPS preconditioning-induced survival.
The 3 isoforms of NOS that are distributed over the brain have distinct biological functions in neuropathological conditions and are selectively induced in different cell types by systemic LPS injection. In the mature brain, high-dose LPS (5 to 50 mg/kg) predominantly upregulated iNOS and downregulated eNOS in blood vessels and glial cells.17 In contrast, medium-dose LPS (2.5 to 10 mg/kg) induced eNOS in astrocytes and blood vessels, but not in neurons.7 Here, we provided evidence that low-dose LPS (0.05 mg/kg) selectively upregulated eNOS, but not iNOS or nNOS, in neurons and endothelial cells in the neonatal brain. The mechanisms for upregulating and downregulating different isoforms of NOS in different neural cell types after LPS remain unknown. They may be related to the ontogenic brain (immature versus mature brain) responses and to the LPS dose administered. Our finding that low-dose LPS selectively upregulated eNOS expression in neurons and endothelial cells in the neonatal brain may be unique and distinct from LPS preconditioning in the mature brain, in which the upregulated expression of both iNOS and eNOS in glial and endothelial cells has been reported.4,7,8,18
eNOS is important in vascular signaling and tone regulation and leukocyte-endothelial interactions. eNOS phosphorylation at serine residue 1177 upregulates eNOS enzymatic activity and NO production.19 eNOS-deficient mice are hypertensive and have an increased infarct volume after a stroke.3,19 In contrast, eNOS transgenic mice in which eNOS is overexpressed in endothelial cells are resistant to stress-induced hypotension, inflammation, and stroke.3,20 In vitro studies also show that eNOS, activated by tumor necrosis factor α through Akt activation in neurons is cytoprotective against neurotoxicity.21 These findings suggest that eNOS expressed in neurons and endothelium is protective and that the mechanisms may involve antiapoptosis.3,21 The interaction among astrocytes, neurons, and endothelial cells may also play a role in the protective effect of preconditioning1; further in vitro study is needed to examine whether the increased eNOS expression in neurons and endothelial cells per se is protective against HI.
PI3K/Akt activation is common after LPS treatment.22 LPS-induced PI3K/Akt is antiapoptotic in endothelial cells.23 Hypoxic preconditioning protected against HI by upregulating PI3K/Akt activity in the neonatal rat brain.9 PI3K/Akt-dependent eNOS phosphorylation is important in the antiapoptotic effect of insulin, estrogen, and statins in stroke.3 PI3K/Akt signaling leading to eNOS activation in endothelial cells, but not in neurons, contributed to the ischemic tolerance of the gerbil hippocampus.11 Our study showed that Akt activation was required for eNOS upregulation in neurons and endothelial cells and for the acquisition of LPS-induced HI tolerance. We hypothesize that activating Akt-eNOS signaling in neurons and endothelial cells using low-dose LPS is both neuroprotective and vasculoprotective in the neonatal brain.
Both nontranscription and transcription mechanisms may be involved in regulating eNOS expression.3 LPS preconditioning is a delayed type of HI tolerance.4–6 Akt-mediated cell survival through eNOS activation and regulation may involve transcriptional regulation and posttranscription modulation mechanisms. Although eNOS phosphorylation was observed as early as 1 hour after LPS, transcriptional regulation of eNOS may be more related to LPS preconditioning-induced tolerance than nontranscriptional mechanisms. The underlying mechanisms on how low-dose LPS activates Akt, and how Akt activation leads to eNOS expression in neurons and vascular endothelial cells, requires additional study.
Combining neuroprotection and vasculoprotection by activating Akt-eNOS may be a viable strategy in treating HI brain injury.12,18 We found that in low-dose LPS preconditioning of the developing rat brain, Akt was protective by upregulating eNOS expression in neurons and endothelial cells. Our study leads us to hypothesize that drugs that mimic the beneficial effects of LPS preconditioning by targeting Akt-eNOS signaling are powerful therapeutics for treating HI injury in the neonatal brain.
We thank Chien-Jung Ho for her skillful assistance with animal preparations and Bill Franke (National Cheng Kung University Research and Development Center) for English editing.
Sources of Funding
This study was supported by grants from the Taiwan National Health Research Institute (NHRI-EX 96,97-9414NI), the National Science Counsel (NSC: 97-2811-B-006-014), and the Center for Gene Regulation and Signal Transduction Research, NCKU.
- Received November 27, 2009.
- Revision received February 26, 2010.
- Accepted March 17, 2010.
Iwase K, Miyanaka K, Shimizu A, Nagasaki A, Gotoh T, Mori M, Takiguchi M. Induction of endothelial nitric-oxide synthase in rat brain astrocytes by systemic lipopolysaccharide treatment. J Biol Chem. 2000; 275: 11929–11933.
Yin W, Signore AP, Iwai M, Cao G, Gao Y, Johnnides MJ, Hickey RW, Chen J. Preconditioning suppresses inflammation in neonatal hypoxic ischemia via Akt activation. Stroke. 2007; 38: 1017–1024.
Hashiguchi A, Yano S, Morioka M, Hamada J, Ushio Y, Takeuchi Y, Fukunaga K. Up-regulation of endothelial nitric oxide synthase via phosphatidylinositol 3-kinase pathway contributes to ischemic tolerance in the CA1 subfield of gerbil hippocampus. J Cereb Blood Flow Metab. 2004; 24: 271–279.
Lee HT, Chang YC, Tu YF, Huang CC. VEGF-A/VEGFR-2 signaling leading to cAMP response element-binding protein phosphorylation is a shared pathway underlying the protective effect of preconditioning on neurons and endothelial cells. J Neurosci. 2009; 29: 4356–4368.
Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates, II ed. San Diego: Academic Press; 1986.
Wung BS, Cheng JJ, Shyue SK, Wang DL. NO modulates monocyte chemotactic protein-1 expression in endothelial cells under cyclic strain. Arterioscler Thromb Vasc Biol. 2001; 21: 1941–1947.
Kunz A, Park L, Abe T, Gallo EF, Anrather J, Zhou P, Iadecola C. Neurovascular protection by ischemic tolerance: role of nitric oxide and reactive oxygen species. J Neurosci. 2007; 27: 7083–7093.
Yamashita T, Kawashima S, Ohashi Y, Ozaki M, Ueyama T, Ishida T, Inoue N, Hirata K, Akita H, Yokoyama M. Resistance to endotoxin shock in transgenic mice overexpressing endothelial nitric oxide synthase. Circulation. 2000; 101: 931–937.
Wong F, Hull C, Zhande R, Law J, Karsan A. Lipopolysaccharide initiates a TRAF6-mediated endothelial survival signal. Blood. 2004; 103: 4520–4526.