SEARCH

SEARCH BY CITATION

Keywords:

  • apoptosis;
  • baicalein;
  • cerebral ischemia;
  • neuroprotection;
  • oxygen and glucose deprivation;
  • PI3K/Akt;
  • PTEN

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

J. Neurochem. (2010) 112, 1500–1512.

Abstract

Recently more evidences support baicalein (Bai) is neuroprotective in models of ischemic stroke. This study was conducted to determine the molecular mechanisms involved in this effect. Either permanent or transient (2 h) middle cerebral artery occlusion (MCAO) was induced in rats in this study. Permanent MCAO led to larger infarct volumes in contrast to transient MCAO. Only in transient MCAO, Bai administration significantly reduced infarct size. Baicalein also markedly reduced apoptosis in the penumbra of transient MCAO rats. Additionally, oxygen and glucose deprivation (OGD) was used to mimic ischemic insult in primary cultured cortical neurons. A rapid increase in the intracellular reactive oxygen species level and nitrotyrosine formation induced by OGD was counteracted by Bai, which is parallel with attenuated cell injury. The reduction of phosphorylation Akt and glycogen synthase kinase-3β (GSK3β) induced by OGD was restored by Bai, which was associated with preserved levels of phosphorylation of PTEN, the phophatase that negatively regulates Akt. As a consequence, Bcl-2/Bcl-xL-associated death protein phosphorylation was increased and the protein level of Bcl-2 in motochondria was maintained, which subsequently antagonize cytochrome c released in cytosol. LY294002 blocked the increase in phospho-AKT evoked by Bai and abolished the associated protective effect. Together, these findings provide evidence that Bai protects neurons against ischemia injury and this neuroprotective effect involves PI3K/Akt and PTEN pathway.

Abbreviations used:
3-NT

3-nitrotyrosine

BAD

Bcl-2/Bcl-xL-associated death protein

Bai

baicalein

DMSO

dimethylsulfoxide

Et

ethidine

GSK3β

glycogen synthase kinase-3β

LDH

lactate dehydrogenase

LOX

lipoxygenase

MCAO

middle cerebral artery occlusion

MTT

3-(4, 5-dimethylthiazole-2-yl)-2,5- dipenyltetrazolium bromide

OGD

oxygen and glucose deprivation

PBS

phosphate-buffered saline

PI3K

phosphatidylinositol 3-kinase

PIP2 and 3

phosphoinisitidylinositol-3,4,5-triphosphate

PTEN

the phosphatase and tensin homolog deleted on chromosome 10

ROS

reactive oxygen species

siRNA

small interfering RNA

TTC

2,3,5-triphenyl tetrazolium chloride

TUNEL

terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling

Stroke remains a leading cause of death and adult disability worldwide. Ischemic stroke accounts for approximately 80% of all strokes (Tsuchiya et al. 1992; Feigin et al. 2003). Ischemic insult produces excessive free radicals which are neurotoxic by inducing apoptotic cell death in neurons. The neurotoxicity mediated by free radicals was supported by reduced ischemic injury following application of antioxidant enzymes and scavengers (Naritomi 2001). Thus, it is recognized that free radical scavengers can act as potential neuroprotective agents.

Scutellaria baicalensis Georgi (Huangqin) has been widely used as antibacterial and anti-inflammatory agents for many centuries in the traditional Chinese herbal medicine. Baicalein (Bai) is the most effective antioxidant among the major flavonoids isolated from the roots of Scutellaria baicalensis. In this regard, it has been reported to scavenge reactive oxygen species (ROS), including superoxide inline image, H2O2, and hydroxyl radicals (Hamada et al. 1993; Hanasaki et al. 1994). Baicalein also has the ability to strongly inhibit iron-dependent lipid peroxidation in microsomes (Gao et al. 1995) and mitochondria (Miyahara et al. 1993; He et al. 2009). Results from our laboratory and others have demonstrated that baicalein has the ability to reduce oxidative damage and exert potent neuroprotective effect in a variety of cell types and animal models including experimental ischemia (van et al. 2006; Lapchak et al. 2007; Liu et al. 2007; Jin et al. 2008), indicating that Bai could be an effective pharmacotherapy for the prevention or treatment of neurodegenerative diseases such as stroke. However, the underlying mechanism has not been well elucidated.

Reactive oxygen species derived from ischemia/reperfusion are associated with phosphatidylinositol 3-kinase (PI3K)/AKT signaling pathway that leads to neuronal survival or death (Taylor and Crack 2004; Crack and Taylor 2005). PI3K/AKT is a major cell survival pathway that has been extensively studied. PI3K/Akt pathway promotes cellular survival by phosphorylating and inhibiting death-inducing proteins, including GSK3, Bcl-2/Bcl-xL-associated death protein (BAD), caspase 9, and forkhead transcription factor like 1 (Datta et al. 1997; Cardone et al. 1998; Brunet et al. 1999; Chan 2004; Woodgett 2005). The activity of AKT depends on the availability of phosphoinisitidylinositol-3,4,5-triphosphate (PIP3), which is generated by the enzyme PI3K. PTEN, the phosphatase and tensin homolog deleted on chromosome 10, was originally identified as a tumor suppressor gene mutated in a large percentage of human cancers. PTEN has a critical role in antagonizing PI3K pathways by dephosphorylates PIP3 and converts it back to PIP2. Thus, PTEN is considered to be a key negative regulator of the PI3K/Akt pathway (Waite and Eng 2002). Considering the key role of PI3K/Akt and PTEN in cell survival in models of neurotoxicity, we sought to determine whether PI3K/Akt and PTEN is involved in the neuroprotective effect of Bai. Two stroke models, one of temporary ischemia with reperfusion at 2 h after stroke onset, and one of permanent ischemia was used to assess whether these effects were dependent on the type of ischemia. In addition, oxygen and glucose deprivation (OGD) in primary cultured neurons was used to mimic ischemic insult in vitro. We have demonstrated that baicalein, a known antioxidant, reduce oxidative stress generated by ischemia/reperfusion and promote cell survival involving PI3K/Akt and PTEN signal pathway.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals and reagents

Sprague-Dawley male rats aged 9 weeks (250 ± 20 g) were used in this study. Postnatal Sprague–Dawley rats (days 0–1) were obtained to culture primary cortical neurons. Animal housing, care, and application of experimental procedures were all carried out in strict accordance with National Institutes of Health Guide for the Care and Use of Laboratory Animals. Baicalein were purchased from Sigma. General reagents were obtained from Sigma-Aldrich (St Louis, MO, USA), unless stated otherwise.

Surgery and drug administration

The middle cerebral artery occlusion (MCAO) rat model was performed using the intraluminal suture technique described by Longa et al. (1989). Rats were anesthetized by 10% chloral hydrate (350 mg/kg). A length of 4.0 monofilament nylon suture (20.0–22.0 mm) with its tip rounded was introduced into the MCA through the left internal carotid artery. For permanent ischemia, the filament was left in place; for transient ischemia, the filament was withdrawn after 2 h. The left femoral artery and vein were exposed and cannulated with PE-50 polyethylene tubing (Fisher Scientific). The arterial catheter served for continuous blood pressure recording and blood gas analysis (AVL 990; Homburg, Germany). Body temperature was maintained at 37°C with a homeothermic blanket. Sham-operated control animals underwent all the surgical procedure except occlusion of the MCA by introducing a short thread. A successful occlusion of the right MCA is achieved when the left forelimb is paretic after surgery. Rats failed to induce the symptom after MCAO were excluded from the groups.

Three groups were randomly assigned to: sham group, MCAO group, and MCAO + Bai group. Baicalein was dissolved in Dimethyl Sulfoxide (DMSO) and administered at the dose of 20 mg/kg intraperitoneally 30 min before and 2 h, 4 h after onset of ischemia. Sham group and MCAO group animals received an equal volume DMSO.

Neurological evaluation

Neurological impairment were assessed by an observer blinded to the identity of treatment and scored as described previously (Huang et al. 1994): 0, normal motor function; 1, flexion of contralateral torso and forelimb upon lifting of the whole animal by the tail; 2, circling to the contralateral side but normal posture at rest; 3, leaning to the contralateral side at rest; 4, no spontaneous motor activity.

Evaluation of ischemic size by 2,3,5-triphenyl tetrazolium chloride (TTC) staining

The ischemic area was evaluated by TTC staining. Briefly, after an overdose of chloral hydrate, rats were killed by decapitation at 22 h of reperfusion. The brains were quickly removed and placed in ice-cold saline for 10 min. The brains were cut into a, b, c, d, and e, five 2-mm coronal slices with a rat brain matrix (Harvard Apparatus, MA, USA) at +3, +1, −1, −3, and −5 mm anterior–posterior from the bregma. Sections incubated in 2% TTC (Sigma Chemical Co.) for 30 min and then immediately fixed in 10% formalin overnight. The infarction area, outlined in white, was measured by image analysis software (HMIAS-2000 Image Analysis System, Champion Medical Imaging Co, Wuhan, China) on the posterior surface of each section. The calculation of infarct volume was performed with: non-infarcted area of the ipsilateral hemisphere/total non-infarcted area (from both the ipsilateral and contralateral hemisphere) to avoid the influence of tissue edema (Swanson and Sharp 1994). The mean volume of brain infarction was calculated by summation of the infarct area in each slice multiplied by the thickness of the slice. Cortical infarct volume was calculated by measuring cortical areas of infarction on a section-by-section basis and subcortical infarct volume equalled the difference between total lesion volume and cortical infarct volume.

Immunohistochemistry

Rats were deeply anesthetized with chloral hydrate at 22 h after reperfusion. After transcardiac perfusion with 250 mL of 4% paraformaldehyde in 0.1 M Phosphate Buffer (PB, pH 7.4), the brains were removed and stored in the same paraformaldehyde solution overnight. Multiple, paraffin-embedded, coronal sections (5 μm thick) were taken from brain. After deparaffinizing, sections from rats in different groups were rehydrated and microwaved thrice for 5 min for antigen retrieval. After 60 min of incubation with 3% normal serum in Phosphate Buffered Saline (PBS), the sections were incubated with primary antibodies overnight at 4°C. Rabbit anti-cleaved caspase 3 1 : 200 (Santa Cruz Inc) antibody was used. After rinsing them with PBS, the sections were treated with a rabbit ABC Kit (Santa Cruz Biotechnology). Application of control serum, instead of the primary antibody, on another section of the same brain provided a negative control for each staining. After washing, sections were incubated for 2 h with horseradish peroxidase-conjugated goat anti-rabbit IgG (1 : 200; Santa Cruz Biotechnology) at 37°C, visualized with 3,3′ - Diaminobenzidine, counter-stained with hematoxylin, dehydrated in ethanol, and coverslipped.

For terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) assay, sections were stained by TUNEL, and the TUNEL positive cells were expressed by fluorescein-dUTP with dNTP and Peroxidase (POD) Horseradish with 3′-diaminobenzidine. The detailed procedures followed manufacturer’s protocol for the in situ Apoptosis Detection Kit (In Situ Cell Death Detection Kit; Roche).

Neuronal culture and treatment

Neurons were isolated as previously described (Ming et al. 2006). Briefly, cerebral cortices were dissected from newborn rats (days 0–1) and incubated in 0.125% trypsin for 30 min at 37°C. Tissues were then triturated with fire-polished glass pipettes and plated on poly-l-lysine-coated plates or dishes at a density of 2.5 × 105~106. Neurons were cultured with Dulbecco’s Modified Eagle Media/F12 (1 : 1) supplemented with 10% fetal bovine serum. Cells were maintained in incubators at 37°C under 5% CO2 atmosphere. After 24 h, the culture medium was changed to Neurobasal medium supplemented with 2% B27 and 2 mM glutamine. Glial growth was suppressed by addition of 5-fluoro-2-deoxyuridine and uridine, yielding cultured cells with 90% neurons as determined by NeuN and glial fibrillary acidic protein (GFAP) staining. The medium were replaced with fresh medium every 3 days. Experiments were performed on days 7–10.

Oxygen-glucose deprivation

Neurons were washed once. During OGD, the medium was changed in a HEPES-buffered glucose-free medium that contained (in mmol/L) 154 NaCl, 5.6 KCl, 2.3 CaCl2, 1.0 MgCl2, 3.6 NaHCO3, 5 HEPES, pH 7.4. OGD was induced by incubating neurons in this medium and placing neurons in an anaerobic chamber with an atmosphere of 95% N2 and 5% CO2 at 37°C. OGD was terminated after 2 h by replacing the glucose-free medium back to Neurobasal medium and further incubating the cultures in the normal cell culture incubator for 22 h or particularly indicated duration of re-oxygenation (Reperfusion, R) in individual experiments. Control cells were treated identically except that they were not exposed to OGD. To observe the effect of baicalein, cortical neurons were pre-treated with baicalein for 2 h before exposure to OGD/R treatment and the solvent DMSO (maximum 0.1% final concentration) served as control. LY294002 (10 μM), a PI3 kinase inhibitor, were pre-treated for 30 min before OGD/R when applied to the neurons.

Cell viability assay

The viability of cells was examined by 3-(4, 5-dimethylthiazole-2-yl)-2,5-dipenyltetrazolium bromide (MTT) assay. Neurons were incubated with indicated concentrations of Bai and subjected to OGD for 2 h followed by 22 h of reoxygenation. MTT was added to a final concentration of 0.5 mg/mL for 4 h before the end of the experiment. The supernatant was removed and 150 μL DMSO was added for 20 min. The MTT optical density values were measured on a microplate reader at 570 and 630 nm wavelength light (Synergy HT; BioTek Instruments, Inc. Winooski, VT).

Cell injury assay

The cell injury was detected by Lactate dehydrogenase (LDH) Measurement. Neurons were incubated with indicated concentrations of Bai and subjected to OGD for 2 h followed by 22 h of reperfusion. LDH release was measured in culture medium using the LDH assay kit (Roche Molecular Biochemicals). Medium (100 μL) was transferred from culture wells to 96-well plates and mixed with 100 μL reaction solution provided in the kit. Optical density was measured at 492 nm 30 min later by utilizing a microplate reader (Synergy HT; BioTek Instruments, Inc.). Background absorbance at 620 was subtracted. The maximal releasable LDH was obtained in each well by 15 min incubation with 1% Triton X-100 at the end of each experiment.

Cell death detection

Cell death was measured by a quantitative sandwich enzyme immunoassay by using mouse monoclonal antibodies directed against DNA and histones (cell death detection ELISAPLUS kit; Roche Applied Science). This assay is used to quantify the DNA fragmentation of mono- and oligonucleosomes in the cytoplasm which indicates apoptotic cell death. After treatment, the samples were placed into a streptavidin-coated microplate and incubated with a mixture of anti-histone-biotin and anti-DNA-peroxidase. During the incubation, nucleosomes were captured via their histone component by the anti-histone-biotin antibody while binding to the streptavidin-coated microplate. Simultaneously, anti-DNA-peroxidase binds to the DNA part of the nucleosomes. After removal of the unbound antibodies, the amount of peroxidase retained in the immunocomplex was photometrically determined with 2, 2′-azinodi-(3 ethylbenzthiazolinesulfonic acid) as the substrate. In this assay, Akt small interfering RNA (siRNA) and control siRNA (ConsiRNA) were purchased from Cell Signalling Technology (Danvers, MA, USA).

Annexin V and phosphatidylinositol (PI) binding staining

The assay of Annexin V and PI binding staining was performed with an Annexin V-FITC Apoptosis Detection Kit according to the manufacture’s instructions (Nanjing Keygen Biotech, Nanjing, China). Briefly, Cells were washed twice with cold PBS, harvested and centrifuged at 200 g for 5 min. Cells were resuspended in 1× binding buffer at a concentration of 1 × 106 cells/mL, 100 μL of the solution were transferred to a 5 mL culture tube, and 5 μL annexin V-FITC and 5 μL PI were added. Cells were gently vortex and incubated for 15 min at 37°C in the dark. Finally, 400 μL of 1× binding buffer was added to each tube. Stained cells were analyzed by FACScan flow cytometer (Becton Dickinson). Data were processed using standard software.

Detection of ROS

Intracellular ROS generation was assessed by dihydroethidium (Molecular Probes, Heidelberg, Germany) which is cell-permeable and oxidized by intracellular superoxide inline image to its fluorescent product, ethidine (Et). Et is intracellularly retained, thus allowing quantitative measurement of the cellular ROS levels (Ahlemeyer et al. 2001). Neurons were incubated with dihydroethidium 10 μM for 20 min at 37°C before the end of the experiment. The cellular fluorescence of Et was analyzed using a dual scanning microplate spectrofluorometer (Synergy HT; BioTek Instruments, Inc).

Western blotting

Cells were washed with PBS and homogenated in lysis buffer [20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerolphosphate, 1 mM Na3VO4, 1 μg/mL aprotinin leupeptin and pepstatin, 1 mM Phenylmethylsulfonyl fluoride (PMSF)]. Cell lysates were sonicated twice for 10 s in an Ultrasonic Dismemberator with output 10% (Model 500; Fisher Scientific) and centrifuged at 14 000 g for 20 min at 4°C. The supernatants were subjected to western blots. Protein concentration was determined by Bicinchoninic Acid (BCA) protein assay. Equal amounts of protein was electrophoresed on sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane. The membranes were blocked with 5% non-fat dry milk in Tris-buffered saline and incubated with different primary antibodies overnight at 4°C. The following antibodies were used in this study: monoclonal antibody PTEN (1 : 1000) and polyclonal antibodies against phospho-PTEN (Ser380/Thr382/Thr383, 1 : 1000), phospho-Akt (Ser473, 1 : 1000) and Akt (1 : 1000, Cell Signalling Technology), phospho-GSK3β(Ser9, 1 : 1000) and GSK3β(1 : 1000, Cell Signalling Technology), rabbit anti-Nitrotyrosine antibody (1 : 1000; Sigma). The membrane was treated with horseradish peroxidase-conjugated secondary antibody for 1 h at 37°C. Blots were developed by ECL (Amersham Biosciences, Piscataway, NJ, USA) procedures according to the manufacturer’s recommendation. β-actin antibody (1 : 3000; Cell Signalling Technology) was used as an internal control.

For detection of Bcl-2 and cytchrome c (1 : 1000; Cell Signaling Technology), cells were fractionated into mitochondrial and cytoplasmic compartments with an ApoAlert cell fractionation Kit (Clontech, Palo Alto, CA, USA) according to the manufacturer’s instructions. Purity of the preparation was assessed by immunoblotting with antibody against mitochondrial marker COXIV (1 : 1000; Cell Signaling).

Immunoprecipitation

Five hundred μg crude extract protein was incubated with anti-Bad polyclonal antibodies with gentle rocking overnight at 4°C. Fifty μL of 50% protein A agarose bead slurry was add and incubate with gentle rocking for 3 h at 4°C. The immune complexes were microcentrifuged for 30 s at 4°C. The pellet was washed five times with 500 μL of 1× cell lysis buffer and resuspended with 40 μL 2× SDS sample buffer. The samples were vortex and microcentrifuged for 30 s. Proteins were eluted from the beads by boiling the samples at 95–100°C for 5 min, then microcentrifuge for 1 min at 14 000 g. Immunoprecipitated proteins were analyzed on 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and immunoblotted with anti-Bad and phospho-Bad (S136) specific antibodies (1 : 1000; Cell Signalling Technology).

Statistical analysis

Results were presented as mean ± SEM. Data were analyzed using spss 10.0 software. Data were analyzed by one-way anova followed the Student–Newman–Keuls test. Differences were considered significant at p < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Bai attenuates neurological impairment and brain damage in ischemic rats

Neurological impairment was assessed by neurological scoring of the degree and duration of forelimb flexion and symmetry of movement and forepaw outstretching, as described in the ‘Materials and methods’. Severe contralateral hemiparesis with marked left forelimb flexion and a tendency of circling movement were observed in MCAO model group at 3 h both in permanent MCAO and transient MCAO. Rats in MCAO + Bai group underwent a significantly amelioration of neurological scores at 3 and 24 h compared with rats in MCAO group of transient MCAO (p < 0.05, n = 12). In permanent MCAO, however, there is only significant difference found at 3 h compared between MCAO and MCAO + Bai group (p < 0.05, n = 11; Fig. 1a).

image

Figure 1.  Effects of baicalein (Bai) on neurological scores and infarct volume in ischemic rats. Permanent middle cerebral artery occlusion (MCAO 24 h) or transient middle cerebral artery occlusion (MCAO 2 h and reperfusion 22 h) were performed on rats as described in ‘Material and methods’. Bai was administered at the dose of 20 mg/kg intraperitoneally 30 min before and 2, 4 h after onset of ischemia. Sham group and MCAO group animals received an equal volume of solvent. (a) Evaluation of neurological scores at 3 and 22 h of permanent MCAO or transient MCAO in rats; (b) Bai administration significantly reduced cortical, subcortical, and total infarct volumes in transient MCAO and no significant difference found between MCAO and MCAO + Bai group in permanent MCAO although there was a tendency of reduced infarct size for rats in MCAO + Bai group (shown as percentage of the non-ischemic hemisphere). Data are expressed as mean ± SEM. n = 11–12. *p < 0.05, **p < 0.01 MCAO group versus MCAO + Bai group.

Download figure to PowerPoint

In this study, we selected 24 h for evaluation of the injury because this time provides maximum infarction for both permanent and transient MCA occlusion (Barone et al. 1992). TTC staining allowed the visualization of the ischemic brain damage, which involved the cortex and the subcortex ipsilateral to MCAO. Permanent MCAO led to larger infarction compared with transient MCAO. Baicalein administration has a tendency to decrease the infarct size in permanent MCAO, however, there is no significant difference found between MCAO and MCAO + Bai group. In contrast, Bai treatment significantly reduced total infarction in transient MCAO (p < 0.05, n = 12). Significant differences between MCAO and MCAO + Bai group were seen both for cortical (p < 0.05, n = 12) and subcortical infarction (p < 0.01, n = 12; Fig. 1b). There is no statistical difference in physiological data in inter-group comparisons at all time points (Table 1).

Table 1.   Physiologica data before MCAO (pre-ischemia), 30 min after MCAO (post-ischemia), and 150 min after MCAO (30 min reperfusion for transient ischemia)
 MCAOBaiPre-ischemia30 min post-ischemia150 min post-ischemia
  1. There were no significant differences between groups at the same time points or within groups at different time points. Bai, Baicalein; MCAO, middle cerebral artery occlusion; Hb, hemoglobin; MAP (mean arterial blood pressure), mitogen-activated.

pHPermanent7.40 ± 0.017.38 ± 0.017.39 ± 0.02
+7.39 ± 0.017.38 ± 0.027.38 ± 0.02
Transient7.40 ± 0.017.39 ± 0.027.39 ± 0.01
+7.38 ± 0.017.38 ± 0.017.40 ± 0.02
PaCO2 (mmHg)Permanent43.1 ± 3.247.3 ± 6.245.4 ± 5.3
+47.5 ± 4.448.7 ± 5.743.4 ± 5.4
Transient43.9 ± 3.649.3 ± 6.642.4 ± 4.3
+44.8 ± 4.148.3 ± 7.141.5 ± 3.7
PaO2 (mmHg)Permanent78.1 ± 5.670.5 ± 9.474.6 ± 7.4
+81.2 ± 7.272.7 ± 7.676.3 ± 9.1
Transient76.3 ± 6.973.1 ± 10.679.1 ± 7.7
+75.4 ± 9.171.7 ± 9.673.2 ± 12.6
MAP (mmHg)Permanent93.6 ± 4.192.7 ± 5.495.4 ± 5.9
+92.8 ± 5.890.8 ± 7.294.3 ± 4.6
Transient91.6 ± 4.794.7 ± 6.193.5 ± 6.5
+95.1 ± 5.895.4 ± 3.291.9 ± 7.9
Temperature (°C)Permanent37.2 ± 0.2437.1 ± 0.4637.3 ± 0.12
+37.2 ± 0.2037.3 ± 0.1837.2 ± 0.27
Transient37.1 ± 0.0937.1 ± 0.1237.2 ± 0.26
+37.1 ± 0.1337.2 ± 0.1637.1 ± 0.21
Hb (g/dL)Permanent15.7 ± 1.515.9 ± 1.216.2 ± 1.3
+15.8 ± 1.416.2 ± 0.815.7 ± 2.1
Transient15.4 ± 0.916.7 ± 1.416.4 ± 1.2
+16.1 ± 1.115.8 ± 0.915.6 ± 1.5

Bai represses caspase 3 activation and apoptosis in transient MCAO rats

The expression of cleaved-caspase 3 and TUNEL staining are shown in Fig. 2(a) and (b). Massive immunoreactivity of cleaved caspase 3 was found mostly in cortical penumbra area in transient MCAO rats. Almost no immunoreactivity or positive TUNEL staining were observed in the sham rats. Bai treatment markedly decreased the immunostaining of cleaved caspase 3 compared with MCAO groups (p < 0.05, n = 5–6). As a marker of apoptosis, the positive TUNEL staining was evident in MCAO group while mild in MCAO + Bai groups (p < 0.05, n = 5–6).

image

Figure 2.  Immunohistochemical change of cleaved-caspase 3 and TUNEL staining in rats subjected to transient middle cerebral artery occlusion (MCAO 2 h and reperfusion 22 h). Baicalein (Bai) was administered at the dose of 20 mg/kg intraperitoneally 30 min before and 2, 4 h after onset of ischemia. Sham group and MCAO group animals received an equal volume of solvent. Representative photographs immunostained with antibodies against cleaved caspase 3 and TUNEL staining showed as (i–iii) and (iv–vi), respectively. Bai treatment repressed caspase 3 activation and apoptotic cell death in transient ischemic rats. Scale bars, i–iii: 20 μm; iv–vi: 50 μm. (a) Quantification analysis of positive expression of cleaved caspase 3; (b) Quantification of apoptotic cells by TUNEL staining. Data are expressed as mean ± SEM. n = 5–6. **p < 0.01 MCAO group versus Sham group; p < 0.05 MCAO + Bai group versus MCAO group.

Download figure to PowerPoint

Bai increases cell viability and prevents cell death induced by OGD/R in primary cultured cortical neurons

MTT assay is an important indicator of mitochondrial function and has been used to quantify cell survival. Primary cultured cortical neurons were subjected to OGD/R treatment as described above. As shown in Fig. 3(a), MTT activity was significantly decreased by OGD/R treatment (p < 0.01). Bai 0.35–3.5 μM significantly increased cell viability reduced by OGD/R treatment (p < 0.05; Fig. 3a).

image

Figure 3.  Effects of Baicalein (Bai) on primary cultured cortical neurons exposure to oxygen and glucose deprivation (OGD) 2 h and reperfusion (R) 22 h. Bai at indicated concentrations applied to neurons 2 h before OGD/R. (a) Cell viability was measured by 3-(4, 5-dimethylthiazole-2-yl)-2,5-dipenyltetrazolium bromide assay. Data were normalized by control as 100% (n = 12 per group); (b) cell injury was measured by lactate dehydrogenase release assay (n = 12 per group); (c) cell apoptosis was measured by Annexin V and PI binding staining using Flow cytometry assay (n = 12 per group). Data are expressed as mean ± SEM. *p < 0.05, **p < 0.01 versus control; p < 0.05 versus OGD/R without Bai.

Download figure to PowerPoint

The release of LDH into the culture media occurs with the loss of plasma membrane integrity, a process most often associated with necrotic cell death. Figure 3(b) shows OGD/R treatment resulted in a significant increase in LDH level compared with control (p < 0.01). Bai 0.35–3.5 μM blocked OGD/R-induced LDH release (p < 0.05; Fig. 3b).

Using the dual staining approach with PI and Annexin V staining detected by flow cytometry, PI (−)/annexin (+) cells represent the early apoptotic populations and PI (+)/annexin (+) cells represent late apoptotic populations. Results are shown in Fig. 3(c). Both PI (−)/annexin (+) and PI (+)/annexin (+) cells were significantly increased when exposed to OGD/R compared with controls (p < 0.01). Bai 3.5 μM significantly decreased the number of apoptotic neurons (p < 0.05).

Bai decreases the intracellular ROS level and nitrotyrosine immunoreactivity induced by OGD/R in primary cultured cortical neurons

The intracellular level of ROS in cortical cultures at 0.5, 2, and 4 h of reperfusion after OGD/R was measured. As shown in Fig. 4(a), a rapid, but transient increase in the intracellular ROS level was observed with a maximum at 0.5 h of reperfusion (p < 0.01 control vs. OGD/R at R-0.5 h). ROS production decreased at 2 h in contrast to 0.5 h after reperfusion (p < 0.05 control vs. OGD/R at R-2 h). When Bai 3.5 μM administered on the neurons subjected to OGD/R, the increased ROS level induced by OGD/R was markedly reduced both at 0.5 and 2 h of reperfusion (p < 0.05 OGD/R vs. OGD/R + Bai at R-0.5 h and R-2h). We further estimated nitrotyrosine immunoreactivity as an index of the level of oxidative and nitrosative stress. As shown in Fig. 4(b) and (c), the 3-nitrotyrosine formation was hardly detected in control neurons while markedly increased at reperfusion 2 h after OGD/R (p < 0.05 control vs. OGD/R) which was reduced by Bai 3.5 μM treatment (p < 0.05 OGD/R vs. OGD/R + Bai). LY294002, the PI3K inhibitor, could mostly abolish the inhibitory effects of Bai on intracellular ROS production and nitrotyrosine formation induced by OGD/R.

image

Figure 4.  Baicalein decreased the intracellular reactive oxygen species (ROS) level and nitrotyrosine immunoreactivity induced by oxygen and glucose deprivation (OGD) 2 h and reperfusion (R) in primary cultured cortical neurons. Baicalein (Bai) 3.5 μM applied to neurons 2 h before OGD/R and LY294002 10 μM applied to neurons 30 min before OGD/R. (A) Intracellular ROS generation was assessed at 0.5, 2 and 4 h of reperfusion by dihydroethidium. Bai decreased the intracellular ROS level, particularly at 0.5 h of reperfusion (n = 12 per group); (b) nitrotyrosine formation was detected by western blot in whole cell lysates at 2 h of reperfusion. A representative of three independent experiments is shown; (c) quantification of nitrotyrosine formation in the western blot indicated a significant reduction by Bai treatment (n = 3). Data are expressed as mean ± SEM. *p < 0.05, **p < 0.01 versus control; p < 0.05 versus OGD/R; p < 0.05 versus OGD/R + Bai.

Download figure to PowerPoint

Effects of Bai on Akt, P-Akt, GSK3β, P-GSK3β, PTEN and P-PTEN induced by OGD/R in primary cultured cortical neurons

As shown in Fig. 5(a) and (b), OGD/R treatment decreased the level of phosphorylation of Akt (p < 0.01 OGD/R vs. Con at R-2h and p < 0.05 at R-22h). However, up-regulated levels of phospho-Akt were detected in the presence of Bai (p < 0.01 OGD/R vs. OGD/R + Bai at R-2h and p < 0.05 at R-22h). The phosphoinositide-3 kinase (PI3K) inhibitor LY294002 mostly abolished Bai-induced phosphorylation of Akt (p < 0.05 OGD/R + Bai vs. OGD/R + Bai + LY at R-2h; Fig. 5a and b).

image

Figure 5.  Effects of baicalein on Akt, P-Akt, glycogen synthase (GSK)3β, P-GSK3β, PTEN, and P-PTEN in primary cultured cortical neurons exposure to oxygen and glucose deprivation (OGD) 2 h and reperfusion (R) 2 or 22 h. Baicalein (Bai) 3.5 μM applied to neurons 2 h before OGD/R and LY294002 10 μM applied to neurons 30 min before OGD/R. Whole cell lysates were used for western blots. (a) Representative blots and quantification analysis of Akt, P-Akt, GSK3β, and P-GSK3β (n = 3); (b) representative western blots and quantitative analysis of PTEN and p-PTEN. Data are expressed as mean ± SEM (n = 3). *p < 0.05, **p < 0.01 versus control; p < 0.05, ††p < 0.01 versus OGD/R; p < 0.05 versus OGD/R + Bai.

Download figure to PowerPoint

Our data mentioned above demonstrates Bai activates Akt likely through PI3K/Akt pathway, which suggests downstream effectors of Akt may contribute to neuroprotection of Bai. The way by which Akt regulates its anti-apoptotic effect is by phosphorylating many effector proteins, including the cytoplasmic protein kinase GSK3β. As shown in Fig. 5(a) and (b), GSK3β phosphorylation was decreased by OGD/R (p < 0.01 OGD/R vs. Con at R-2h and R-22h). The addition of Bai reversed the decreased P-GSK3β (p < 0.05 OGD/R vs. OGD/R + Bai at R-2h and R-22h). Inhibition of PI3K/Akt pathway with LY294002 reduced Bai-induced GSK3β phosphorylation (p < 0.05 OGD/R + Bai vs. OGD/R + Bai + LY at R-22h; Fig. 5a and b).

It has been suggested that PTEN is constitutively active in cells. Its activity could be regulated by phosphorylation-dependent modulation of the protein stability. It is therefore of our interest to examine the protein level and phosphorylation status of PTEN after OGD/R treatment. The results of western blots demonstrated that the protein levels of PTEN and Akt were not changed while phosphorylation of PTEN decreased after OGD/R (p < 0.01 OGD/R vs. Con at R-2h and p < 0.05 at R-22h), which can be reversed by the presence of Bai (p < 0.05 OGD/R vs. OGD/R + Bai at R-2h). LY294002 can mostly abolish this effect of Bai (p < 0.05 OGD/R + Bai vs. OGD/R + Bai + LY at R-2h; Fig. 5c and d).

Suppression PI3/Akt pathway antagonize neuroprotective effect of Bai

Our studies demonstrate that Bai activates the pro-survival factor PI3K/Akt pathway. Confirming those results, we have shown here that knockdown Akt by Akt siRNA or inhibition PI3K by LY294002 could attenuate the effects of Bai promoting neuronal survival after OGD/R insult. As shown in Fig. 6(a), Akt knockdown was confirmed by immunoblotting. Consistent with data described above, Bai-treated neurons showed significant decreased cell death induced by OGD/R (p < 0.05 OGD/R vs. Con and p < 0.05 OGD/R vs. OGD/R + Bai) by the MTT cell viability assay. Further, this effect was blunted by Akt knockdown or PI3K inhibition (p < 0.05 OGD/R + Bai vs. OGD/R + Bai + AktsiRNA or OGD/R + Bai + LY; p < 0.05 OGD/R + Bai + ConAktsiRNA vs. OGD/R + Bai + AktsiRNA). Serving as respective controls, ConsiRNA, AktsiRNA ,or LY294002 failed to modulate OGD-induced cell death (Fig. 6b). Results obtained from the MTT assay were further confirmed by an ELISA that quantifies mono- and oligonucleosomal fragmented DNA in the cytoplasm of the apoptotic neurons. In parallel to the MTT results, Bai treatment significantly decreased cell death induced by OGD/R, AktsiRNA, or LY294002 blunted this response (Fig. 6c). These results reconfirm that the neuroprotective effect of Bai was mediated by PI3K/Akt pathway.

image

Figure 6.  Suppression PI3/Akt pathway impairs neuroprotective effect of Bai on primary cultured cortical neurons exposure to oxygen and glucose deprivation (OGD) 2 h and reperfusion (R) 22 h. Baicalein (Bai) 3.5 μM applied to neurons 2 h before OGD/R and LY294002 10 μM applied to neurons 30 min before OGD/R. (a) Knockdown of Akt by Akt small interfering RNA (siRNA) was confirmed by immunoblotting. β-actin served as a loading control (n = 3). (b) Cell viability was assessed by 3-(4, 5-dimethylthiazole-2-yl)-2,5-dipenyltetrazolium bromide assay. Knockdown Akt by Akt siRNA or inhibition phosphatidylinositol 3-kinase (PI3K) by LY294002 attenuated the enhanced cell viability by Bai after OGD/R insult. Data were normalized by control as 100% (n = 12 per group). (c) Cell death was assessed by quantifying mono- and oligonucleosomal fragmented DNA in the cytoplasmic extracts by ELISA. Knockdown Akt by Akt siRNA or inhibition PIK3 by LY294002 blocked the reduced cell death by Bai after OGD/R (n = 12/group). *p < 0.05, **p < 0.01 versus control; p < 0.05 versus OGD/R; p < 0.05 versus OGD/R + Bai; p < 0.05 OGD/R + Bai + AktsiRNA versus OGD/R + Bai + control siRNA.

Download figure to PowerPoint

Bai affects of phosphorylation of BAD, preserves Bcl-2 from the mitochondria and prevents cytochrome c release induced by OGD/R in primary cultured cortical neurons

To investigate the downstream signaling molecules of Akt, BAD was immunoprecipitated and followed by the analysis of its phosphorylation status of serine residues 136 (S136). At 22 h reperfusion after OGD, the reduced phosphorylation of BAD at S136 was increased by Bai (p < 0.05 OGD/R vs. Con and p < 0.01 OGD/R vs. OGD/R + Bai). The effect of Bai on phosphorylation of BAD was dependent on Akt-mediated signaling, because LY294002 abrogated Bai-induced phosphorylation (p < 0.05 OGD/R + Bai vs. OGD/R + Bai + LY; Fig. 7a). To determine the effect of Bai on anti-apoptotic protein, we analyzed mitochondrial Bcl-2. An induction of Bcl-2 occurred by Bai treatment. OGD/R and inhibition of PI3 kinase by LY294002 induced the loss of Bcl-2 from the mitochondria (p < 0.05 OGD/R vs. Con; p < 0.05 OGD/R vs. OGD/R + Bai; p < 0.05 OGD/R + Bai vs. OGD/R + Bai + LY; Fig. 7b), which indicate that Bai preserved OGD/R-induced loss of Bcl-2 from the mitochondria via PI3K/Akt pathway. In consistent with this, there was also a significant increase in cytosolic cytochrome c at 22 h reperfusion after OGD. Bai blocked OGD-induced cytochrome c release, which was prevented by LY294002 (p < 0.05 OGD/R vs. Con; p < 0.05 OGD/R vs. OGD/R + Bai; p < 0.05 OGD/R + Bai vs. OGD/R + Bai + LY; Fig. 7c).

image

Figure 7.  Baicalein (Bai) preserves protein levels of Bcl-2 on mitochondria, inhibits cytochrome c release in cytosol and increases phosphorylation of Bcl-2/Bcl-xL-associated death protein (BAD) on primary cultured cortical neurons exposure to oxygen and glucose deprivation (OGD) 2 h and reperfusion (R) 22 h. Bai 3.5 μM applied to neurons 2 h before OGD/R and LY294002 10 μM applied to neurons 30 min before OGD/R. (a) Representative western blot and quantitative analysis of phosphorylation of BAD and BAD in whole cell lysates (n = 3); (b) representative western blot and quantitative analysis of Bcl-2 in mitochondrial fraction, COX-served as a loading control (n = 3); (c) representative western blot and quantitative analysis of cytochrome c in cytosolic fraction, β-actin served as a loading control (n = 3). Data are expressed as mean ± SEM. *p < 0.05 versus control; p < 0.05, ††p < 0.01 versus OGD/R; p < 0.05 versus OGD/R + Bai.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The brain is particularly susceptible to oxidative stress because it consumes a large quantity of oxygen. Although oxygen is necessary for life, it paradoxically produces ROS as a by-product of its metabolism. Under normal conditions, the rate of free radicals formation is equal to that of their elimination. However, during ischemia and reperfusion, this balance is perturbed either because of increased free radicals production or decreased activity of cellular defense systems (Valko et al. 2007), which lead to free radicals massively accumulate in the brain. Those radicals are redox active and have the potential to react with and damage nearby cellular targets including lipids, proteins and DNA, which is followed by more delayed post-ischemic inflammation and apoptosis, and these events are involved in the progression and expansion of brain injury (Peters et al. 1998; Hata et al. 2000). Thus, oxidative stress are believed to contribute to neuronal loss after ischemia/reperfusion (Andersen 2004; Margaill et al. 2005) and the scavenger of free radicals is considered to be important for achieving neuroprotection against ischemia/reperfusion injury.

Arachidonic acid metabolism is a potential source for ROS generation during ischemia and reperfusion. As a result of the activation of phospholipases in responding to brain ischemia, increased arachidonic acid release occurs (Katsuki and Okuda 1995) and is metabolized by three enzyme systems: cyclooxygenase, LOX, and epoxygenase. Metabolism of arachidonic acid by these enzymes produce free radicals and peroxides (Siesjo and Katsura 1992; Paller and Jacob 1994; Phillis et al. 2006). Lipoxygenase (LOX), one of the major metabolic enzymes, catalyzes the incorporation of molecular oxygen into specific positions of arachidonic acid. Based on the position of oxygen insertion, LOX is classified as 5-, 12-, or 15-LOX (Shimizu and Wolfe 1990). 12-LOX is the major LOX found in the brain. Blockage 12/15-LOX activation prevents ROS generation and accumulation during cerebral ischemia. The natural product Bai has been identified as a specific inhibitor of 12/15-LOX. As a polyphenol which belongs to the flavone subgroup, Bai also has superior free radical scavengering and antioxidant effects because of its o-trihydroxyl structure in the A ring (Gao et al. 1999) and lipophilic character (Saija et al. 1995). Gao et al. reported Bai scavenges hydroxyl radical, DPPH (1,1-Diphenyl-2-Picrylhydrazyl) radical, and alkyl radical in a dose-dependent manner in vitro, inhibits lipid peroxidation of rat brain cortex mitochondria induced by Fe2•−ascorbic acid, 2,2′-azobis (2-amidino-propane) dihydrochloride (AAPH), or NADPH, protects human neuroblastoma SH-SY5Y cells against H2O2-induced injury (Gao et al. 1999). Our previous studies showed that Bai inhibited ischemic neuronal cell loss related to its antioxidant action (Liu et al. 2007). In this study, a rapid but transient increase in the intracellular ROS level was observed shortly after OGD. As we know, the major oxidative stress produced by superoxide is derived from its reaction with nitric oxide yielding peroxynitrite (ONOO), a highly reactive species and much more toxic than inline image, to elicit nitrosative damage. A characteristic reaction of ONOO is the nitration of protein-bound tyrosine residues. Neurons are capable of generating ONOO because of their capacity to simultaneously produce inline image and NO. Our results also demonstrated increment of 3-nitrotyrosine (3-NT) formation produced by OGD/R treatment. Excessive ROS and 3-NT plays an important role in induction of apoptotic death in cultured cortical neurons. When Bai was applied to neurons exposure to OGD/R, ROS, and 3-NT formation was markedly decreased accompanied by increased cell survival. This study was reported for the first time that Bai protects cortical neuronal cells from OGD, the most often used in vitro model mimicking metabolic aspects of ischemia, induced cell death by inhibiting apoptosis, which greatly extends previous reports on the effects of Bai on systemic tissues. In agreement with this, our investigation in vivo demonstrated that Bai administration significantly reduced infarct size and apoptosis induced by transient MCAO and reperfusion in rats, which is consistent with the previous reports (onso-Galicia et al. 1999; van et al. 2006). In contrast to permanent MCAO, reperfusion prompts a short-lasting steep maximum to nearly fivefold increase of ROS concentrations, while permanent vessel occlusion led to a more gradual increase by about twofold levels at 3-h post-occlusion (Peters et al. 1998). Deleterious effect of superoxide radicals is enhanced by reperfusion and is a leading cause in reperfusion injury. Thus, suppression ROS plays a more predominant role in rescuing cells from reperfusion injury. However, the restoration of flow makes the therapeutic agent to reach the ischemic tissue to minimize infarct development. These reasons may explain the better outcome for Bai administration in transient MCAO compared with permanent MCAO.

PI3K/Akt regulates the survival response against oxidative stress-associated neuronal apoptosis (Dudek et al. 1997; Hong et al. 2001; Manning and Cantley 2007). In the CNS, decreased Akt activity has been linked to the neuronal death induced by ischemia or hypoxia (Hirai et al. 2003, 2004; Luo et al. 2003). In contrast, increased Akt activity contributes to the neuroprotection induced by hypothermia (Zhao et al. 2005) or hypoxic preconditioning (Zhang et al. 2007; Wang et al. 2008). Constitutive activation of Akt signaling is sufficient to block cell death induced by a variety of apoptotic stimuli. Phosphorylation of Akt (Ser473) is required for Akt activation. Several downstream targets of Akt have been recognized as apoptosis-regulatory molecules. Activated Akt promotes cell survival and suppresses apoptosis by phosphorylation and inhibition of several downstream substrates, including GSK3β (Cross et al. 1995), a mechanism by which neurons are proposed to become resistant to apoptotic stimuli (Hetman et al. 2000). BAD is also the target of Akt. As a proapoptotic member of the Bcl-2 family, BAD promote cell death by binding and neutralizing the function of anti-apoptotic Bcl-2. Phosphorylation of BAD promotes binding to 14-3-3 proteins to be sequestered in the cytosol and to prevent an association between BAD with Bcl-2 (Yang et al. 1995; Zha et al. 1996; Datta et al. 2000). Akt inhibit the apoptotic activity of BAD by phosphorylation of BAD at Ser136 (Datta et al. 1999). Following cell stress, the loss of Akt activity leads to BAD dephosphorylation and translocation to mitochondria, where it binds with Bcl-2 and activates the mitochondrial cell death pathway to release cytochrome c into cytosol. In this study, we demonstrated that Bai treatment enhanced the reduced Akt phosphorylation after OGD/R. Pharmacologic inhibition PI3K or silencing Akt expression by siRNA impaired the ability of Bai to protect against OGD/R-induced cortical neurons death. Moreover, Bai increased phosphorylation of GSK3βand BAD, and inhibited OGD/R-induced loss of Bcl-2 from mitochondria. Cytochrome c release in cytosol was sequentially blocked. These are consistent with the results that baicalein inhibit caspase 3 activation and apoptotic death in transient MCAO rats. In addition, pre-treatment with LY294002, the specific inhibitor of PI3K, blocked baicalein to increase Akt phosphorylation and the following targets when exposure to OGD/R. We also found the decreased intracellular ROS level by baicalein after OGD/R was abolished in the presence of LY294002, further verified that PI3K/Akt-mediated neuroprotection regulated oxidative stress response (Di et al. 2006).

The biological effects of Akt are determined by the balance between the activity of PI3K and PTEN. PTEN is a major negative regulator of the PI3K/Akt signaling pathway. The phosphorylation of three specific residues Ser380/Thr382/383 of PTEN is required for its biological activity (Torres and Pulido 2001). Unlike AKT, PTEN phosphorylation results in its inactivation, not activation. Dephosphorylation of PTEN increases PTEN activity and reduces PIP3 availability leading to dephosphorylation of AKT. PTEN has been demonstrated to act as an important mediator of ROS production and of mitochondria-dependent apoptosis (Zhu et al. 2006). Over-expression of PTEN increased the sensitivity of hippocampal neurons to excitotoxicity (Gary and Mattson 2002), whereas knockdown PTEN or pharmacological down-regulation of PTEN phosphorylation protected brain tissue from ischemic damage (Ning et al. 2004; Hong et al. 2006). So, we further examined the possibility that PTEN is involved in the effect of Bai on OGD/R-treated cortical neurons. Consistent with this prediction, we found that PTEN is rapidly dephosphorylated after OGD and this dephosphorylation is reversed by Bai. Therefore, it is likely that Bai scavenge free radicals and suppress oxidant stress in ischemia/reperfusion, which causes PTEN to lose its activity and, in turn, increase AKT activity and inhibit the downstream mediated apoptotic cell death.

To summarize, we have demonstrated that baicalein administration have neuroprotective effect against ischemic brain injury both in vivo and in vitro which is related to its ability to scavenge free radicals and mediated by PI3K/Akt and PTEN pathway. The findings provide further insight into the mechanisms through which Bai exerts its beneficial effect on the injured brain and suggests Bai may be a promising agent for clinic therapy of ischemic brain damage.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported by the grants from the National Science Fund for Distinguished Young Scholars of China (No. 30425024), the National Basic Research Program of China (973 Program; No. 2007CB507404) and the Chang Jiang Scholars Program to Dr Jian-Guo Chen, the Joint Research Fund for Overseas Chinese Young Scholars to Dr Yong Xia and Dr Jian-Guo Chen (No. 30728010).

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References