Two models for determining the mechanisms of damage and repair after hypoxic-ischaemic injury in the developing rat brain

Hypoxia-ischemia and therapeutic hypothermia in the neonatal mouse brain--a longitudinal study

Therapeutic hypothermia is standard of care for infants with hypoxic ischemic encephalopathy. Murine models of hypoxic-ischemic injury exist; however, a well-established mouse model of therapeutic hypothermia following hypoxic-ischemic injury is lacking. The goal of this study was to develop a full-term-equivalent murine model of therapeutic hypothermia after hypoxia-ischemia and examine magnetic resonance imaging, behavior, and histology in a region and sex specific manner. Hypoxic-ischemic injury was induced at postnatal day 10 in C57BL6 mice using a modified Vannucci model. Mice were randomized to control, hypothermia (31˚C for 4h), or normothermia (36˚C) following hypoxic-ischemic injury and stratified by sex. T2-weighted magnetic resonance imaging was obtained at postnatal day 18 and 30 and regional and total cerebral and cerebellar volumes measured. Behavioral assessments were performed on postnatal day 14, 21, and 28. On postnatal day 18, normothermic mice had smaller cerebral volumes (p < 0.001 vs. controls and p = 0.009 vs. hypothermia), while at postnatal day 30 both injured groups had smaller volumes than controls. When stratified by sex, only normothermia treated male mice had smaller cerebral volumes (p = 0.001 vs. control; p = 0.008 vs. hypothermia) at postnatal day 18, which persisted at postnatal day 30 (p = 0.001 vs. control). Female mice had similar cerebral volumes between groups at both day 18 and 30. Cerebellar volumes of hypothermia treated male mice differed from control at day 18, but not at 30. Four hours of therapeutic hypothermia in this murine hypoxic-ischemic injury model provides sustained neuroprotection in the cerebrum of male mice. Due to variable degree of injury in female mice, response to therapeutic hypothermia is difficult to discern. Deficits in female behavior tests are not fully explained by imaging measures and likely represent injury not detectable by volume measurements alone.

Growth hormone receptor immunoreactivity is increased in the subventricular zone of juvenile rat brain after focal ischemia: a potential role for growth hormone in injury-induced neurogenesis

BACKGROUND

During recovery from an ischemic brain injury, a cerebral growth hormone (GH) axis is activated. Whilst GH has been demonstrated to be neuroprotective both in vitro and in vivo, a role for GH in neuro-restorative processes after brain injury has yet to be studied.

OBJECTIVE

To explore a role for GH in injury-induced neurogenesis by examining GH receptor (GH-R) immunoreactivity within the subventricular zone (SVZ) of juvenile rats after brain injury and by testing the proliferative capacity of GH on embryonic mouse neural stem cells.

DESIGN

Twenty-one day old rats were subjected to unilateral hypoxic-ischemia of the brain and sacrificed 1-15days later. Coronal brain sections from these animals and age-matched naïve controls were immunostained for GH-R and cell markers of neurogenesis. The level of GH-R immunoreactivity in the ipsilateral and contralateral SVZ of each animal was semi-quantified both by independent blinded scoring by two examiners and blinded image analysis. To examine the effect of GH on proliferation of embryonic mouse neural stem cells, cells were treated with increasing concentrations of rat pituitary GH for 48h in the presence of 5'-bromo-2'-deoxyuridine.

RESULTS

The level of GH-R immunoreactivity in the ipsilateral SVZ was significantly increased 5days after injury vs. the contralateral SVZ, coinciding both spatially and temporally with injury-induced neurogenesis. The population of GH-R immunopositive cells in the ipsilateral SVZ at this time was found to include proliferating cells (Ki67 immunopositive), neural progenitor cells (nestin immunopositive) and post-proliferative migratory neuroblasts (doublecortin immunopositive). Stimulation of embryonic mouse NSCs with physiological concentrations of rat pituitary GH elicited a dose-dependent proliferative response.

CONCLUSION

These results indicate a novel role for GH and its receptor in injury-induced neurogenesis, and suggest that GH treatment may potentiate endogenous neuro-restorative processes after brain injury.

Phosphorus magnetic resonance spectroscopy 2 h after perinatal cerebral hypoxia-ischemia prognosticates outcome in the newborn piglet

Phosphorus magnetic resonance spectroscopy ((31)P MRS) often reveals apparently normal brain metabolism in the first hours after intrapartum hypoxia-ischemia (HI) at a time when conventional clinical assessment of injury severity is problematic. We aimed to elucidate very-early, injury-severity biomarkers. Twenty-seven newborn piglets underwent cerebral HI: (31)P-MRS measures approximately 2 h after HI were compared between injury groups defined by secondary-energy-failure severity as quantified by the minimum nucleotide triphosphate (NTP) observed after 6 h. For severe and moderate injury versus baseline, [Pi]/[total exchangeable high-energy phosphate pool (EPP)] was increased (p < 0.001 and < 0.02, respectively), and [NTP]/[EPP] decreased (p < 0.03 and < 0.006, respectively): severe-injury [Pi]/[EPP] was also increased versus mild injury (p < 0.04). Mild-injury [phosphocreatine]/[EPP] was increased (p < 0.004). Severe-injury intracellular pH was alkaline versus baseline (p < 0.002). For severe and moderate injury [total Mg]/[ATP] (p < 0.0002 and < 0.02, respectively) and [free Mg] (p < 0.0001 and < 0.02, respectively) were increased versus baseline. [Pi]/[EPP], [phosphocreatine]/[Pi] and [NTP]/[EPP] correlated linearly with injury severity (p < 0.005, < 0.005 and < 0.02, respectively). Increased [Pi]/[EPP], intracellular pH and intracellular Mg approximately 2 h after intrapartum HI may prognosticate severe injury, whereas increased [phosphocreatine]/[EPP] may suggest mild damage. In vivo(31)P MRS may have potential to provide very-early prognosis in neonatal encephalopathy.

Proteolytic activity during cortical development is distinct from that involved in hypoxic ischemic injury

Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases involved in brain development and the etiology of adult cerebral injuries. In this study, we determined the MMP-2 and 9 responses following hypoxic ischemia (HI) injury in the developing brain. First, we characterized the developmental changes of MMP activity in the rat brain from embryonic day 18 (E18) to postnatal day 120 (P120). MMP-2 activity was high from E18 to P3 and decreased with age (P< or =0.001), while MMP-9 activity was not detectable. MMP-2 immunoreactivity was closely associated with differentiating cortical plate and subplate neurons. Next, we characterized the proteolytic changes after unilateral HI brain injury in 3- (P3) and 21- (P21) day-old rats. Zymography revealed that in the P21 rat brain, MMP-9 activity (150 and 92 kDa forms) was increased at 6 h and remained elevated 24 h post-injury in the ipsilateral injured hemisphere (P< or =0.001), whereas there was a gradual increase in MMP-2 (65 kDa) activity, reaching a peak at 5 days (P< or =0.001). Similarly, quantitative real time polymerase chain reaction (qRT-PCR) indicated significant elevations in MMP-9 and MMP-2 mRNA expression in the injured cortex (P< or =0.05) and hippocampus (P< or =0.05) at 1 and 5 days post-injury, respectively in the P21 rat brain. In the P3 rat brain, zymography results revealed that both pro (92 kDa) and cleaved (87 kDa) MMP-9 activities were upregulated in the ipsilateral injured hemisphere from 6 h to 1 day after injury (P< or =0.001). In contrast, cleaved MMP-2 (60 kDa) was only moderately upregulated at 6 h (P< or =0.01), while pro MMP-2 (65 kDa) levels were unaffected. MMP-9 mRNA expression was also increased at 6 h (P< or =0.05) following injury at P3, whereas MMP-2 expression remained unchanged compared with the uninjured contralateral hemisphere. Immunohistochemistry indicated that MMP-9 protein expression was localized predominantly to neurons and peri-vascular astrocytes in the affected regions at early time points, whereas MMP-2 was present on reactive astrocytes surrounding the infarct at later time points. Together, these results indicate that MMP-2 may be primarily associated with the development and differentiation of cortical plate neurons and wound recovery processes. Conversely, MMP-9 appeared to be associated with more acute processes during the period of lesion development.

Distinct neuronal growth hormone receptor ligand specificity in the rat brain

A cerebral growth hormone axis is activated during recovery from brain injury and centrally administered growth hormone can rescue injured neurons. It remains unclear, however, whether this treatment effect occurs directly via neuronal growth hormone receptors. Immunohistochemistry confirmed growth hormone receptor protein on neuronal cell bodies in the rat cortex. Surprisingly, we found that central treatment with bovine growth hormone, which is equipotent to rat growth hormone in the rat periphery, failed to rescue cortical neurons following hypoxic ischemic injury. We further investigated the actions of rat and bovine growth hormone on primary neuron-enriched cultures of fetal rat cortex. In agreement with the in vivo treatment studies, rat but not bovine growth hormone rescued neurons from nutrient deprivation-induced cell death (p<0.05). This neuroprotective effect was inhibited by the selective growth hormone receptor antagonist G120D (p<0.001). Furthermore, rat but not bovine growth hormone had trophic effects on uninjured cultures (p<0.001). Immunocytochemistry showed growth hormone receptor on neurons within the neuron-enriched cultures. We show for the first time that the protective and trophic effects of rat growth hormone are mediated via growth hormone receptors on neurons and that the rodent neuronal growth hormone receptor exhibits unique ligand specificity.

Prolactin is involved in glial responses following a focal injury to the juvenile rat brain

A cerebral growth hormone axis is activated following brain injury in the rat and treatment with growth hormone is neuroprotective. We have now investigated whether the closely related prolactin axis has similar properties following injury to the developing rat brain. From one day following a unilateral hypoxic ischemic injury, prolactin immunoreactivity was increased in the affected cortex parallel to the development of the injury (P<0.001). Initial prolactin and prolactin receptor staining on penumbral neurons progressively decreased whereas astrocytes remained strongly immunopositive. Reactive microglia also became strongly prolactin immunoreactive. Unlike growth hormone, central treatment with prolactin failed to rescue neurons in this paradigm. This was confirmed in vitro; rat prolactin failed to protect neurons under conditions for which growth hormone was neuroprotective. However, prolactin had trophic and pro-proliferative effects on glia (P<0.001). We confirmed the expression of the prolactin receptor in vitro by reverse transcriptase polymerase chain reaction, and show its strong association with astrocytes as compared with neurons by immunocytochemistry. In summary, we show for the first time that hypoxia ischemia induces a robust activation of the prolactin axis in regions of the cerebral cortex affected by injury. The lack of neuroprotective properties in vivo and in vitro indicates that, unlike growth hormone, prolactin is not directly involved in neuronal rescue in the injured brain. Its strong relation to glial reactions and its gliatrophic effects suggest that the prolactin axis is primarily involved in a gliogenic response during recovery from cerebral injury.

Minocycline modulates chemokine receptors but not interleukin-10 mRNA expression in hypoxic-ischemic neonatal rat brain

Hypoxic-ischemic (HI) brain injury in the perinatal period causes significant morbidity. Minocycline (MN) is a tetracycline derivative that has reduced brain injury in various animal models of neurodegeneration, including perinatal ischemia. To determine whether MN can modulate the expression of chemokine receptors and interleukin-10 (IL10) in a model of neonatal brain injury, we produced an HI insult to the right cerebral hemisphere (ipsilateral) of the 7-day-old rat (PD7) by right common carotid artery ligation and 2.25 hr of hypoxia in 8% oxygen. MN (45 mg/kg, i.p.) or vehicle (PBS) was injected twice: 2 days and immediately before the HI insult. At 0, 1, 3, and 24 hr and 14 days after HI, total RNA from the ipsilateral and contralateral (exposed to hypoxia only) hemispheres was extracted, reverse transcribed, and amplified with gene-specific primers using a semiquantitative RT-PCR for macrophage inflammatory protein-1alpha), interferon-inducible protein (IP-10), C-C chemokine receptor 5 (CCR5; MIP-1alpha receptor), C-X-C chemokine receptor 3 (CXCR3; IP-10 receptor), and IL10. We found that, in the ipsilateral hemisphere, a significant (P < 0.05) increase in MIP-1alpha, IP-10, CCR5, and CXCR3 mRNA levels was observed. MN treatment decreased mRNA levels for CCR5 and CXCR3. In contrast, the levels of antiinflammatory cytokine IL10 were markedly decreased as a result of HI insult. Treatment with MN, however, had no effect on IL10. We conclude that MN decreased proinflammatory chemokine receptor expression but had little or no influence on the expression of antiinflammatory cytokine IL10. These effects confirm the antiinflammatory effect of MN in neonatal HI brain injury.

Development of a postnatal 3-day-old rat model of mild hypoxic-ischemic brain injury

Improvements in both obstetric and paediatric care have been responsible for a continuing reduction in mortality in extremely premature infants. However, higher survival rates have been at the expense of more long-term neurological damage. Various animal models have been developed to study the effect of hypoxic-ischemic insults on the brain. However, established models like the postnatal day 7 rat model represent damage found in term infants rather than in preterm infants of 24-28 weeks' gestation, and produce a severe form of injury resulting in high mortality rates. In this study we developed a reliable model of minor hypoxic-ischemic brain injury in postnatal day 3 rats. At this maturity, the pattern of damage represents that expected in a preterm infant suffering a non-lethal perinatal insult. We found that minor changes in duration of insult and both temperature and humidity produced wide fluctuations in the degree of injury observed. By maintaining strict control over experimental conditions including duration of insult, temperature and humidity, we produced a reliable model of minor injury primarily affecting all five areas of the cerebral cortex, and also the thalamus (area 7) and basal ganglia (area 8). Differences were significant compared to normal controls and sham-operated animals (p<0.05). These areas represent the primary motor, insular, visual and temporal cortices. The overall mortality rate in this study was 12.3%.

Insulin-like growth factor-1 and post-ischemic brain injury

Insulin-like growth factor-1 (IGF-1) is a naturally occurring neurotrophic factor that plays an important role in promoting cell proliferation and differentiation during normal brain development and maturation. The present review examines recent evidence that endogenous IGF-1 also plays a significant role in recovery from insults such as hypoxia-ischemia and that giving additional exogenous IGF-1 can actively ameliorate damage. It is now well established that neurons and other cell types die many hours or even days after initial injury due to activation of programmed cell death pathways. IGF-1 and its binding proteins and receptors are intensely induced within damaged brain regions following brain injury, suggesting a possible a role for IGF-1 in brain recovery. Exogenous administration of IGF-1 within a few hours after brain injury is now known to be protective in both gray and white matter and leads to improved somatic function. In contrast, pre-treatment is ineffective, likely reflecting limited intracerebral penetration of IGF-1 into the uninjured brain. The neuroprotective effects of IGF-1 are mediated by IGF-1 receptors and its binding proteins and are specific to particular cellular phenotypes and brain regions. The window of opportunity for treatment with IGF-1 is limited to a few hours after normothermic brain injury, reflecting its specific actions on early, intracellular events in the apoptotic cascade. However, injury-associated mild post-hypoxic hypothermia, which delays the development of cell death, can shift and dramatically extend the window of opportunity for delayed treatment with IGF-1. Such a combined approach is likely to be essential for any clinical treatment.

Selective cortical alteration after hypoxic-ischemic injury in the very immature rat brain

Distinctive cerebral lesions with disruptions to the developing white matter are found in very low birth weight (VLBW) infants. Although hypoxia-ischemia (HI) is a causal pathway, the pathogenesis of cerebral white matter injury in the VLBW infant is not fully understood. Pertinent murine models would facilitate the investigation of the processes leading to these cerebral lesions and enable the evaluation of therapeutic strategies. Postnatal d 3 (P3) rats are at a stage of cortical oligodendroglial maturation and axonal outgrowth similar to very preterm infants. Our aim was to characterize the effects of a focal hypoxic-ischemic injury at P3 on subsequent cerebral development. Three groups of P3 Wistar rats were investigated: group I underwent right carotid ligation followed by 6% hypoxia for 30 min (HI), group 2 had carotid ligation only, and group 3 had no intervention. At P21, in the HI group, the right cortical area was reduced compared with controls (p < 0.01). There were no significant alterations in the size of the dorsal hippocampus, striatum, and thalamus. The cortical myelinated area was reduced in the HI animals compared with controls (p < 0.01). There was a corresponding loss of myelinated axons extending up into the cortex, with deep cortical neuronal and axonal architecture markedly disrupted. Glial fibrillary acidic protein immunohistology showed a reactive gliosis in the deep parietal cortex (p < 0.01). Moderate HI injury in the immature rat brain compromised cortical growth and led to a selective alteration of cortical myelinated axons with persistent gliosis. These alterations induced at P3 by unilateral HI share neuropathological similarities with the diffuse white matter lesions found in VLBW infants.

A delayed increase in hippocampal proliferation following global asphyxia in the neonatal rat

Adult neurogenesis has been shown to be upregulated following a wide variety of brain injury paradigms. During the first weeks of postnatal life there is around 50 fold more neurogenesis occurring than in the adult CNS, yet little is known regarding the effect of neonatal brain injury on this developmental proliferation. We have investigated the effect of a global perinatal birth asphyxia on postnatal proliferation at 2, 5, 8, 11, 15, 21 and 28 days after birth (injury) using a 3H-thymidine tracer study. We found a specific upregulation of proliferation at 5 days after the injury within the injured hippocampus only, with an associated increase in hippocampal mass and without any changes in GFAP content at any timepoint. Perinatal asphyxia did not alter proliferation within the cerebellum, sub ventricular zone, olfactory bulb, cervical or thoracic spinal cord. Similarly, no changes in corticosterone levels were induced by the injury. Since there were no changes in GFAP content we hypothesize that this increased proliferation is likely neurogenetic, similar to what is seen in the adult brain following injury. Further we show that the dramatic increase in corticosterone at the end of the stress hyporesponsive period is not responsible for the equally dramatic decrease in postnatal proliferation within the CNS.

RAGE is expressed in pyramidal cells of the hippocampus following moderate hypoxic-ischemic brain injury in rats

The receptor for advanced glycation end products (RAGE) is a multi-ligand member of the immunoglobulin superfamily of cell surface molecules. The RAGE-ligand interaction has a putative role in a range of chronic disorders and is also known to contribute to both inflammatory/degenerative processes as well as regeneration in peripheral nerve injury. We have investigated the expression of RAGE in the moderate hypoxic-ischemic (HI) rat brain injury model in order to determine if this receptor is involved in the cellular perturbation mediated by ischemic stress. RAGE mRNA levels were detected by in situ hybridization using a DIG-labelled 48 mer oligonucleotide probe. Results showed a high level of expression of RAGE mRNA in the CA1/2 pyramidal cell layer of the hippocampus on the lesioned side of the brain 72 h after a moderate hypoxic-ischemic insult. RAGE was not expressed on the control side of the hippocampus. The RAGE-positive cells had a unique morphology, being angular in shape and atrophied with a condensed cell nucleus. They were NeuN-positive and were identified as dying cells by staining with thionin/acid fuchsin. A subset of cells was positive for cleaved Caspase-3, a marker for apoptosis. Together these data show that RAGE is expressed in dying neurons and suggest that RAGE may have a role in neuronal cell death mediated by ischemic stress. Identification of the ligand for RAGE in the ischemic brain may lead to a better understanding of RAGE-mediated cellular dysfunction in the CNS.

Magnesium pre-treatment reduces neuronal apoptosis in newborn rats in hypoxia-ischemia

Hypoxic-ischemic brain damage has significant mortality and morbidity in newborns. Although the role of magnesium in neonatal hypoxic-ischemic brain injury related to N-methyl-D-aspartate receptors has been widely studied; the effects of magnesium on neuronal apoptosis have not been known exactly in hypoxia-ischemia. The aim of this study was to investigate the effects of magnesium on neuronal apoptosis in the 7-day-old rat hypoxia-ischemia model. Seven-day-old rats were administered magnesium sulfate (group 1; n=9) or saline (group 2; n=9) intraperitoneally before hypoxia-ischemia. Additionally 18 seven-day-old rats were given magnesium sulfate (group 3; n=9) or saline (group 4; n=9) after hypoxic-ischemic insult. Neuronal apoptosis was investigated by the dUDP-biotin nick end-labeling (TUNEL) method following 3-day recovery in all subjects. In evaluating TUNEL-positive cells, we firstly calculated the areas (mm(2)) of brain regions, hippocampus, striatum, cortex, in right and left hemispheres in subjects by IMAGE analysis. The numerical density was calculated as the number of cells per square millimeter by counting all TUNEL-positive cells. Afterwards, the ratio of right side numeric density to sum of right and left side numeric densities (right Apoptosis Index) was calculated for every brain region in rats receiving magnesium and compared to vehicle groups. The right Apoptosis Index of the hippocampus in magnesium pre-treated rats (mean+/-S.D.; 36.6+/-22.1) was significantly lower than vehicle (61.0+/-16.0; P<0.05); whereas right apoptosis indices were not changed by magnesium pre-treatment in striatum and cortex. Additionally, magnesium sulfate administration following hypoxic-ischemic insult also had no effect on right apoptosis indices in all three brain regions. It is concluded that magnesium might have a role in preventing neuronal apoptosis due to neonatal hypoxic-ischemic brain injury.

Expression of the p75 neurotrophin receptor by striatal cholinergic neurons following global ischemia in rats is associated with neuronal degeneration

The induction of the p75 neurotrophin receptor (p75NTR) on striatal cholinergic neurons by global hypoxic-ischemia has been reported to promote neuron survival. We have found, however, while the p75NTR-expressing neurons survive the insult for the first 5 days, subsequently they undergo shrinkage, loss of choline acetyl transferase (ChAT) expression, and more than 96% are eventually lost by 8 days. In contrast ChAT-expressing cells in the surrounding region of the infarction, do not express p75NTR and there is no evidence of neuronal loss. These results suggest the expression of p75NTR on cholinergic interneurons of the rat striatum is associated with delayed neuronal degeneration.

Minocycline markedly protects the neonatal brain against hypoxic-ischemic injury

Hypoxic-ischemic brain injury in the perinatal period is a major cause of morbidity and mortality. Presently, there are no proven effective therapies with which to safeguard the human neonatal brain against this type of injury. Minocycline, a semisynthetic tetracycline, has been shown to be neuroprotective in certain adult ischemic injury/stroke and neurodegenerative disease models. However, minocycline's neuroprotective effects have not been assessed after insults to the neonatal brain. We now report that minocycline administered either immediately before or immediately after a hypoxic-ischemic insult substantially blocks tissue damage in a rodent model of neonatal hypoxic-ischemic brain injury. Minocycline treatment prevents the formation of activated caspase-3, a known effector of apoptosis, as well as the appearance of a calpain cleaved substrate, a marker of excitotoxic/necrotic cell death. To our knowledge, this is the first report of a systemic treatment that can be administered after a hypoxic-ischemic insult, which provides robust, nearly complete neuroprotection to the developing brain. Our data suggest that minocycline or a related neuroprotective tetracycline may be a candidate to consider in human clinical trials to protect the developing brain against hypoxic-ischemic-induced damage.

Neuroprotective effects of the N-terminal tripeptide of IGF-1, glycine-proline-glutamate, in the immature rat brain after hypoxic-ischemic injury

Insulin growth factor 1 (IGF-1) has an important role in brain development and is strongly expressed during recovery after a hypoxic-ischemic injury. Some of its central actions could be mediated through the N-terminal tripeptide fragment of IGF-1: Gly-Pro-Glu (GPE). The neuroprotective properties of GPE given after a moderate injury in the developing rat brain were evaluated and the binding sites of [(3)H]GPE characterised by autoradiography. After right unilateral injury, GPE or vehicle (V) was injected in the right lateral ventricle (i.c.v.) or in the peritoneal cavity (i.p.) of 21-day-old rats. The percentage of surviving neurons in CA1-2 of the hippocampus was higher in the animals treated with 30 microg of GPE i.c.v. (V: 7.7+/-4.9%, GPE: 26.4+/-7.5%, P=0.02) and 300 microg i.p. (V: 30.2+/-9.1%, GPE: 68.8+/-10.6%, P=0.02) than in animals receiving vehicle. I.p. injection of 300 microg of GPE (V: 78.4+/-7.5%, GPE: 88.4+/-3.2%, P=0.04) was also neuroprotective in the lateral cortex. I.c.v. injection of [(3)H]GPE suggested binding to glial cells in the white matter tracts, the cortex and striatum as opposed to neurons. Although the precise mode of action of GPE is unknown, this study suggests that local administration of GPE is neuroprotective after brain HI injury via glial cells. In addition, systemic administration of GPE showed a more widespread neuroprotective effect. GPE may represent a complementary pathway for central and systemic IGF-1's antiapoptotic effects.

Growth hormone as a neuronal rescue factor during recovery from CNS injury

There is growing evidence to suggest that growth hormone plays a role in the growth and development of the CNS. Specifically, growth hormone has been implicated in promoting brain growth, myelination, neuronal arborisation, glial differentiation and cognitive function. Here we investigate if growth hormone has a role in the recovery from an unilateral hypoxic-ischaemic brain injury. Using moderate (15 min hypoxia) and severe (60 min hypoxia) models of hypoxic-ischaemia in juvenile rats and standard immunohistochemical techniques, we found intense growth hormone-like immunoreactivity present within regions of cell loss by 3 days (P<0.05). Growth hormone-like immunoreactivity was observed on injured neurones, myelinated axons, glial cells within and surrounding infarcted tissue and on the choroid plexus plus ependymal cells within the injured hemisphere. The pattern of immunoreactivity suggests that (a) growth hormone (or a growth hormone-like substance) is transported via the cerebrospinal fluid and (b) that growth hormone (or a growth hormone-like substance) is acting in a neurotrophic manner specifically targeted to injured neurones and glia. To test this hypothesis we treated a moderate hypoxic-ischaemic brain injury with 20 microg of rat growth hormone by intracerebroventricular infusion starting 2 h after injury (n=12/group). After 3 days the animals were killed and the extent of neuronal loss quantified. Growth hormone treatment reduced neuronal loss in the frontoparietal cortex (P<0.001), hippocampus (P<0.01) and dorsolateral thalamus (P<0.01) but not in the striatum. This spatial distribution of the neuroprotection conveyed by growth hormone correlates with the spatial distribution of the constitutive neural growth hormone receptor, but not with the neuroprotection offered by insulin-like growth factor-I treatment in this model. These results suggest that some of the neuroprotective effects of growth hormone are mediated directly through the growth hormone receptor and do not involve insulin-like growth factor-I induction.In summary, we have found that a growth hormone-like factor increased in the brain in the days after injury. In addition, treatment with growth hormone soon after an hypoxic-ischaemic injury reduced the extent of neuronal loss. These results further suggest that a neural growth hormone axis is activated during recovery from injury and that this may act to restrict the extent of neuronal death.

Perinatal hypoxia-/ischemia-induced endothelin-1 mRNA in astrocyte-like and endothelial cells

Under pathological conditions in the adult CNS, such as ischemia, subarachnoid hemorrhage and Alzheimer's disease, endothelin (ET)-1- and -3-like immunoreactivities are elevated in astrocytes of the injured adult brain. However, it is not clear whether this is due to increased synthesis or increased binding of ET-1. Further, it is not known whether ET-1 expression is altered in the perinatal brain after cerebral hypoxia/ischemia (H/I). Here, we determined the sites of ET-1 expression in perinatal mouse brain after H/I injury by in situ hybridization using a probe specific for the ET-1 gene. Astrocyte-like cells, which do not normally express ET-1 mRNA, showed high levels of ET-1 mRNA expression. Endothelial cells of the capillaries and small vessels also showed an increased level of ET-1 mRNA. Our data suggest that ET-1 mRNA levels in the astrocyte-like cells and vascular endothelial cells are dynamically regulated by ischemia and may participate in perinatal ischemia-related neural damage.

Hypoxic-ischemic brain injury in the newborn: pathophysiology and potential strategies for intervention

There is increasing clinical and experimental data describing the evolution of hypoxic-ischemic encephalopathy in the perinatal period. Outcome to the fetus is determined not only by the impact of gross asphyxial insult, but also external factors that sensitize the brain to injury. Delayed neuronal and glial death occurring in the hours and days after the insult by apoptotic and related processes are observed following severe injury, and offer the most promise for pharmacological intervention. Furthermore, new technologies allow the identification of subtle insults with evolving encephalopathies that have implications for long-term neurological outcome. Application of this knowledge will allow us to identify strategies for early intervention and prevent the course of damage caused by hypoxic-ischemic injury.

PU.1 expression in microglia

The transcription factor PU.1 has a pivotal role in both the generation and function of macrophages. To determine whether PU.1 is also involved in microglial regulation, we investigated its expression following hypoxic-ischemia (HI) brain injury and in the BV-2 microglial cell line. We found that microglia constitutively expressed high levels of PU.1 protein in both their 'resting' and 'activated' states.

Spectrophotometric measurement of experimental brain injury

Freshly sampled brain tissue exposed to 2,3,5-triphenyltetrazolium chloride (TTC) acquires a red color because mitochondrial enzymes reduce the colorless TTC to a red, water-insoluble formazan deposit. Pan-necrotic areas remain uncolored, which enables quantitation of experimental brain injury by optical scanning and image analysis of serial slices to determine the relative volume of red versus infarcted, non-stained, tissue. The accuracy of this method can be challenged, however, when infarction is accompanied by areas of partial, scattered injury where differences in coloration are difficult to see or quantify. We tested the feasibility of measuring scattered injury using a principle which underlies standard assays for in vitro cell survival, namely extracting deposited formazan with a solvent and measuring its level by spectrophotometry. Anesthetized, adult Sprague Dawley rats were subjected to 12 min of cerebral ischemia to produce selective, delayed neuronal death in hippocampus, striatum and cortex. Some rats also received 6 h of whole-body hypothermia treatment (31.5-32.5 degrees C) immediately after ischemia. Ischemia rats and non-operated controls were sacrificed 1 week later. Hippocampus and portions of cerebrum were incubated 90 min in a 2% TTC solution and then soaked in a measured volume of 50:50 ethanol and dimethylsulfoxide to extract the red formazan product. Spectrophotometric measurements of the extract showed a diminished formazan coloration (absorbance/g brain) in all samples from the untreated ischemia group compared to non-operated controls. This apparent brain injury was attenuated in the group of ischemia rats that received hypothermia treatment. We conclude that solvent extraction and spectrophotometric quantitation of formazan has potential utility as an objective way to index experimental brain injury even if this is diffuse in nature and not amenable to measurement by conventional image analysis techniques.

Expression of the activin axis and neuronal rescue effects of recombinant activin A following hypoxic-ischemic brain injury in the infant rat

Neurotrophic factors are induced in the brain in response to injury and may restrict the extent of neuronal loss and facilitate recovery. We have previously reported a strong neuronal induction of activin betaA subunit mRNA expression after a hypoxic-ischemic (HI) injury in the rat brain. Here, we further extended our studies to examine a role for the activin inhibitory binding protein, follistatin after injury and also to determine the potential of activin as a neuronal rescue agent. Ribonuclease protection assay (RPA) was used to quantify the time course of the mRNA expression of activin betaA subunit and follistatin, following a 60-min HI brain injury. Activin betaA subunit mRNA level increased in the contralateral hemisphere 5 h after injury and returned to normal at 10 h post injury. In contrast, follistatin mRNA levels decreased in the same hemisphere at 5 and 10 h after injury. The effect of intracerebroventrically (i. c.v.) administered recombinant human activin A or its antagonist, inhibin A, on neuronal death after a 15-min HI brain injury was determined for a number of brain regions. One microgram activin A (n=23) reduced the neuronal loss in the hippocampal CA1/2 region, dorsolateral striatum but not in the parietal cortex. In contrast, 1 microg of inhibin A (n=18) did not have a significant effect on the extent of neuronal loss in any of the affected regions. This pattern of neuroprotection was consistent with the distribution of immunoreactivity for the activin receptor type II subunit. These results demonstrate that activin A, but not its functional antagonist inhibin A, can enhance the survival of injured hippocampal and striatal neurons. Since follistatin is thought to exert a neutralising effect on activin A activity, the down-regulation of follistatin expression post injury may be allowing activin A to become more accessible to neurons after injury. Overall, these results suggest a role of the activin axis in modulating the survival of specific populations of injured neurons.

Alterations in the neural growth hormone axis following hypoxic-ischemic brain injury

Recently, there has been considerable interest in determining the role of the growth hormone receptor (GHR) in the central nervous system (CNS). The aim of this study was to investigate the role of circulating growth hormone (GH) and the neural GHR after hypoxic-ischemic (HI) brain injury in the 21-day old rat. We observed growth hormone receptor/binding protein (GHR/BP) immunoreactivity to be rapidly upregulated following a severe unilateral HI injury. There was a biphasic increase with an initial rise occurring in blood vessels within a few hours after injury followed by a secondary rise evident by 3 days post-hypoxia in microglia/macrophages, some of which are destined to express insulin-like growth factor-I (IGF-I). There was also an increased immunoreactivity in reactive astrocytes, some of which were in the process of dividing. Subsequently, we attempted to activate the endothelial GHR/BP which was found to be increased after injury by treating with 15 microgram g-1 day-1 s.c. bGH for 7 days. Circulating concentrations of IGF-I fell after injury and were restored with GH treatment (P=0.001), whereas treatment of normal animals had no effect on serum IGF-I. Peripheral GH treatment increased the cerebrospinal fluid (CSF) concentration of immunoreactive IGF-I in the injured rats (P=0.017). GH treatment also reversed the systemic catabolism caused by the injury but had no significant neuroprotective effects. These results indicate that GH therapy can be used to reverse the systemic catabolism that occurs after CNS injury. In addition, these data suggest a role for the neural GHR during the recovery from brain injury, both in terms of the induction of IGF-I and in terms of glial proliferation.

Neuronal death and survival in two models of hypoxic-ischemic brain damage

Two unilateral hypoxic-ischemia (HI) models (moderate and severe) in immature rat brain have been used to investigate the role of various transcription factors and related proteins in delayed neuronal death and survival. The moderate HI model results in an apoptotic-like neuronal death in selectively vulnerable regions of the brain while the more severe HI injury consistently produces widespread necrosis resulting in infarction, with some necrosis resistant cell populations showing evidence of an apoptotic type death. In susceptible regions undergoing an apoptotic-like death there was not only a prolonged induction of the immediate early genes, c-jun, c-fos and nur77, but also of possible target genes amyloid precursor protein (APP751) and CPP32. In contrast, increased levels of BDNF, phosphorylated CREB and PGHS-2 were found in cells resistant to the moderate HI insult suggesting that these proteins either alone or in combination may be of importance in the process of neuroprotection. An additional feature of both the moderate and severe brain insults was the rapid activation and/or proliferation of glial cells (microglia and astrocytes) in and around the site of damage. The glial response following HI was associated with an upregulation of both the CCAAT-enhancer binding protein alpha (microglia only) and NFkappaB transcription factors.

Melanocortin-4 receptor messenger RNA expression is up-regulated in the non-damaged striatum following unilateral hypoxic-ischaemic brain injury

Melanocortin peptides (alpha-melanocyte-stimulating hormone, adrenocorticotropin and fragments thereof) have been shown to have numerous effects on the central nervous system, including recovery from nerve injury and retention of learned behaviour, but the mechanism of action of these peptides is unknown. A family of five melanocortin receptors have recently been discovered, two of which (melanocortin-3 and melanocortin-4 receptors) have been mapped in the rat brain. We have tested the hypothesis that the expression of one or more of the messenger RNAs for three melanocortin receptors (melanocortin-3, melanocortin-4 and melanocortin-5 receptors) would be altered in rat brain following unilateral transient hypoxic-ischaemic brain injury. In this study, using in situ hybridization, we show that melanocortin-4 receptor messenger RNA was up-regulated in the striatum in the non-damaged hemisphere within 24 h after severe hypoxic-ischaemic injury compared with control brains (P<0.05). In a small group of animals, this induction was not blocked by treatment with the anticonvulsant, carbamazepine. Expression of melanocortin-3 receptor messenger RNA in the brain was not altered in this hypoxic-ischaemic injury model and melanocortin-5 receptor messenger RNA was not detected in either control or hypoxic-ischaemic injured rat brains. We hypothesize that the up-regulation of melanocortin-4 receptor messenger RNA expression in the contralateral striatum may be involved in transfer of function to the uninjured hemisphere following unilateral brain injury.

Activity and injury-dependent expression of inducible transcription factors, growth factors and apoptosis-related genes within the central nervous system

This review primarily discusses work that has been performed in our laboratories and that of our direct collaborators and therefore does not represent an exhaustive review of the current literature. Our aim is to further discuss the role that gene expression plays in neuronal plasticity and pathology. In the first part of this review we examine activity-dependent changes in the expression of inducible transcription factors (ITFs) and neurotrophins with long-term potentiation (LTP) and kindling. This work has identified particular ITFs (Krox-20 and Krox-24) and neurotrophin systems (particularly the brain-derived neurotrophic factor (BDNF)/tyrosine receptor kinase-B, Trk-B system) that may be involved in stabilizing long-lasting LTP (i.e. LTP3). We also show that changes in the expression of other ITFs (Fos, Jun-D and Krox-20) and the BDNF/trkB neurotrophin system may play a central role in the development of hippocampal kindling, an animal model of human temporal lobe epilepsy. In the next part of this review we examine changes in gene expression after neuronal injuries (ischemia, prolonged seizure activity and focal brain injury) and after nerve transection (axotomy). We identify apoptosis-related genes (p53, c-Jun, Bax) whose delayed expression selectively increases in degenerating neurons, further suggesting that some forms of neuronal death may involve apoptosis. Moreover, since overexpression of the tumour-suppressor gene p53 induces apoptosis in a wide variety of dividing cell types we speculate that it may perform the same function in post-mitotic neurons following brain injuries. Additionally, we show that neuronal injury is associated with rapid, transient, activity-dependent expression of neurotrophins (BDNF and activinA) in neurons, contrasting with a delayed and more persistent injury-induced expression of certain growth factors (IGF-1 and TGFbeta) in glia. In this section we also describe results linking ITFs and neurotrophic factor expression. Firstly, we show that while BDNF and trkB are induced as immediate-early genes following injury, the injury-induced expression of activinA and trkC may be regulated by ITFs. We also discuss whether loss of retrograde transport of neurotrophic factors such as nerve growth factor following nerve transection triggers the selective and prolonged expression of c-Jun in axotomized neurons and whether c-Jun is responsible for regeneration or degeneration of these axotomized neurons. In the last section we further examine the role that gene expression may play in memory formation, epileptogenesis and neuronal degeneration, lastly speculating whether the expression of various growth factors after brain injury represents an endogenous neuroprotective response of the brain to injury. Here we discuss our results which show that pharmacological enhancement of this response with exogenous application of IGF-1 or TGF-beta reduces neuronal loss after brain injury.

CCAAT-enhancer binding protein alpha is expressed in activated microglial cells after brain injury

Microglial cells play important roles in brain injury and repair and are implicated in diseases such as Alzheimer's disease, Creutzfeldt-Jacob disease, multiple sclerosis, the Aids Dementia Complex and stroke. Despite their importance in neuropathology, the underlying molecular basis for the activation of microglia after brain injury is not understood. We show, using RT-PCR, in situ hybridisation, immunocytochemistry, and electrophoretic mobility shift assay, that the CCAAT-enhancer binding protein alpha (C/EBP alpha), a sequence specific DNA-binding protein, is induced in microglial cells, but not astrocytes or neurons, after hypoxic-ischemic brain injury. These results suggest that C/EBP alpha might regulate gene expression and consequentially have a role in the activation and/or proliferation of microglia following brain injury.

Co-ordinated and cellular specific induction of the components of the IGF/IGFBP axis in the rat brain following hypoxic-ischemic injury

Insulin-like growth factor 1 (IGF-1) is induced after hypoxic-ischemic (HI) brain injury, and therapeutic studies suggest that IGF-1 may restrict delayed neuronal and glial cell loss. We have used a well-characterised rat model of HI injury to extend our understanding of the modes of action of the IGF system after injury. The induction of the IGF system by injury was examined by in situ hybridization, immunohistochemistry, Northern blot analysis, RNase protection assay and reverse transcriptase-polymerase chain reaction (RT-PCR). IGF-1 accumulated in blood vessels of the damaged hemisphere within 5 h after a severe injury. By 3 days, IGF-1 mRNA was expressed by reactive microglia in regions of delayed neuronal death, and immunoreactive IGF-1 was associated with these microglia and reactive astrocytes juxtaposed to surviving neurones surrounding the infarct. Total IGF-1 receptor mRNA was unchanged by the injury. IGFBP-2 mRNA was strongly induced in reactive astrocytes throughout the injured hemisphere, and IGFBP-3 and IGFBP-5 mRNA were moderately induced in reactive microglia and neurones of the injured hippocampus, respectively. IGFBP-6 mRNA was induced in the damaged hemisphere by 3 days and increased protein was seen on the choroid plexus, ependyma and reactive glia. In contrast, insulin II was not induced. These results indicate cell type-specific expression for IGF-1, IGFBP-2,3,5 and 6 after injury. Our findings suggest that the IGF-1 produced by microglia after injury is transferred to perineuronal reactive astrocytes expressing IGFBP-2. Thus, modulation of IGF-1 action by IGFBP-2 might represent a key mechanism that restricts neuronal cell loss following HI brain injury.

Asphyxial brain injury--the role of the IGF system

Transient neural injuries, such as asphyxia, can trigger considerable delayed neuronal death. Inappropriate induction of apoptosis is thought to play an important role in this process. Our studies have shown marked changes in the IGF system in the brain in response to these injuries with an induction of insulin growth factor (IGF)-1 and insulin growth factor binding protein (IGFBP)-2 and IGFBP-3 in glial cells in the region of injury. This suggests that the IGF-1 system may be an endogenous neuroprotective system. Earlier administration of IGF-1 - 2 h after injury reduced the phase of secondary neuronal loss suggesting that IGF-1 may well have therapeutic potential as a neuronal rescue agent. The action of IGF-1 appears to involve binding proteins, transport to the site of injury and the IGF-1 receptor and inhibition of apoptosis, but might also involve generation of GPE which itself appears to be neuroprotective. Together these results indicate considerable potential of these agents to treat stroke, perinatal asphyxia and other forms of acute brain injury.

Prostaglandin H synthase-2 and cytosolic phospholipase A2 in the hypoxic-ischemic brain: role in neuronal death or survival?

The breakdown of membrane phospholipids and subsequent arachidonic acid metabolism to prostanoids is a well-documented brain response to cerebral ischemia. To further elucidate the components of this signal transduction pathway, immunocytochemistry was used to determine the levels of two potentially important enzymes, cytosolic phospholipase A2 (cPLA2) and prostaglandin H synthase-2 (PGHS-2), in the immature rat brain following moderate unilateral hypoxic-ischemia (HI). The CA1 pyramidal cells of the hippocampus which undergo delayed neuronal death on the injured side following HI demonstrated a significant induction of PGHS-2 immunoreactivity 48 h post-insult. However, a consistent increase in PGHS-2 was also evident in the resistant dentate granule cells at an earlier time point. Although PGHS-2 is present in both susceptible and resistant cell populations following HI, the possibility remains that divergence further down-stream in the pathway is responsible for selective vulnerability. In contrast to the neuronal PGHS-2 expression, cPLA2 immunoreactivity appears to be of glial origin with increases in and around the CAI-2 pyramidal cell layer at the 72-168-h time points. These results suggest that prostanoids are likely to serve important roles in HI brain damage and repair in infant brain.

Expression of Fos, Jun, and Krox family proteins in Alzheimer's disease

Apoptosis is an active process of cell death characterized by distinct morphological features and is often the end result of a genetic program of events, i.e., programmed cell death (PCD). There is growing evidence supporting a role for apoptosis and/or PCD in Alzheimer's disease (AD), based on DNA fragmentation studies and recent findings of increased levels of inducible transcription factors (ITFs) such as c-Jun in AD brains. We have characterized the expression of a large range of ITFs (c-Fos, Fos B, Fos-related antigens, c-Jun, Jun B, Jun D, Krox20, and Krox24) using multiple antisera in AD postmortem hippocampi and compared this with human control hippocampi as well as Huntington's disease hippocampi and human epilepsy biopsy tissue. We found little evidence of nuclear expression of any ITF except c-Jun in the human postmortem tissue, compared with nuclear staining in biopsy tissue. We found some evidence for increased levels of c-Jun and Krox24 protein and krox24 mRNA in the CA1 region of AD hippocampi, suggesting that PCD may be involved in the pathogenesis of AD. In general, staining characteristics of ITFs varied with different antisera directed against the same protein, indicating the need for caution when interpreting results.

Carotid artery and jugular vein ligation with and without hypoxia in the rat

A continuing concern about the use of extracorporeal membrane oxygenation (ECMO) is the cannulation of the common carotid artery or the internal jugular vein. The authors investigated the changes that might occur in the brain with neck vessel ligation in the normal and the hypoxic rat. Two groups of 60 rats each were studied. The first group was divided into three subgroups of 20 animals each. Subgroup 1 (HH) was hypoxic both 24 hours before and 24 hours after operation. Subgroup 2 (HN) (the ECMO model) was hypoxic before operation and recovered for 24 hours in room air. Subgroup 3 (NN) underwent the entire procedure in room air. For each oxygen environment, four different operations were performed: carotid artery ligation, jugular vein ligation, carotid artery and jugular vein ligation, and dissection of the vessels without ligation (sham). Thus each subgroup was further divided into four sub-subgroups based on the operation performed. Rats were again anesthetized after a 24-hour recovery period and killed using low, blunt cervical dislocation. In the first group of 60 rats, the skull was opened and the brain was carefully removed from the cranial vault and placed in a fixative. The brains were placed in a small magnetic resonance imaging (MRI) head coil in groups of five and scans were obtained to provide T1 and T2 images that correlated with histological sections. MRI scans were reviewed in random, blinded fashion by an imager unaware of how these animals had been treated. The brains were then sectioned coronally at six corresponding levels: frontal, mid and posterior cerebrum, midbrain, pons, and medulla. Histological examination was performed in blinded fashion. The number of lesions (usually ischemic as noted by a decrease in the number of neurons) was totaled for each area of the brain. There were no differences that were consistent or statistically significant in the MR images of brains removed from the head, although it would appear that rats with jugular vein and carotid artery ligation were relatively protected. In the HN group jugular vein ligation was worst, and adding carotid artery ligation was best. In the histological studies the NN group had significantly more lesions than the HH group (P < .01). The second group of 60 rats was divided and treated as the first group in all respects except that MRI was conducted immediately after death on intact heads, and no histological studies were performed. This was done to control for lesions that might have been produced by removal of the brains from the skulls. In this group all findings were right sided. One animal in the HN group showed midcerebral white matter edema after jugular and carotid ligation. Focal anterior cerebral edema was seen in another animal (HH) after isolated carotid ligation. An occipital infarct was found in one animal (HH) after both carotid and jugular ligation. The authors conclude that neck vessel ligation in the hypoxic or normoxic rat causes only occasional and sporadic brain injury much as is seen clinically in newborn ECMO patients.

Marked age-dependent neuroprotection by brain-derived neurotrophic factor against neonatal hypoxic-ischemic brain injury

Hypoxic-ischemic brain injury in survivors of perinatal asphyxia is a frequently encountered clinical problem for which there is currently no effective therapy. Neurotrophins, such as brain-derived neurotrophic factor (BDNF), can protect responsive neurons against cell death in some injury paradigms. While the role of BDNF in hypoxic-ischemic brain injury is not clear, evidence suggests that BDNF may have different effects in the developing, as opposed to the adult, brain. We found that a single intracerebroventricular (ICV) injection of BDNF resulted in rapid and robust phosphorylation of trk receptors in multiple brain regions in the postnatal day (PD) 7 rat brain. BDNF also markedly protected against hypoxic-ischemic brain injury at PD7. It protected against 90% of tissue loss due to hypoxic-ischemia when given just prior to the insult and against 50% of tissue loss when give after the insult. In contrast, ICV injection of BDNF in PD21 and adult rats resulted in little trk phosphorylation and less dramatic protection against unilateral hypoxic-ischemic injury at PD21. Because of its potent neuroprotective actions in the developing brain, BDNF may be a potential treatment for asphyxia and other forms of acute injury in the perinatal period.

Bax expression in mammalian neurons undergoing apoptosis, and in Alzheimer's disease hippocampus

Recent studies indicate that the proto-oncogene Bax, and other related proteins (eg Bcl-2) may play a major role in determining whether cells will undergo apoptosis under conditions which promote cell death. Increased expression of Bax has been found to promote apoptosis, while over-expression of Bcl-2 can inhibit apoptosis. To investigate the role of Bax in nerve cell death in the rat brain we examined the level of Bax expression in cells undergoing apoptosis, using a hypoxic-ischemic stroke model. We found that Bax was expressed at high levels in the nuclei of neurons in the hippocampus, cortex, cerebellum, and striatum on the control side, and that Bax levels increased in hippocampal neurons undergoing apoptosis on the stroke side, and then declined (correlating with cell loss). In the Alzheimer's disease hippocampi we found a concentrated localisation of Bax in senile plaques, which correlated with the localisation of beta-amyloid protein in adjacent sections from the same brains. beta-Amyloid positive plaques are thought to contribute to the Alzheimer's disease process, possibly via an apoptotic mechanism, and this may occur via an increase in Bax in these areas. Bax was also strongly stained in tau-positive tangles in Alzheimer's disease hippocampi, suggesting Bax may play a role in tangle formation. In addition, we observed a loss of Bax expression in the dentate granule cells of Alzheimer's disease hippocampi compared with moderate Bax expression in control hippocampi, and this loss may be related to the survival of these neurons in Alzheimer's disease. Finally, we observed substantially different staining patterns of Bax using three different commercially available antisera to Bax, indicating the need for caution when interpreting results in this area.

Loss of Ref-1 protein expression precedes DNA fragmentation in apoptotic neurons

Ref-1 is a bifunctional protein that has been implicated in the transcriptional regulation of AP-1 elements and in DNA repair. To investigate whether Ref-1 is involved in programmed cell death its expression was measured in the 21-day-old rat brain at various time-points following a moderate unilateral hypoxic-ischemic (HI) insult. The CA1 pyramidal cells, which are selectively vulnerable to HI injury, showed a significant decrease in Ref-1 immunoreactivity 48 h-7 days post-insult. This loss of Ref-1 immunoreactivity may contribute to a decrease in endogenous repair activity and the development of apoptosis in the CA1 pyramidal cells.

Relation between delayed impairment of cerebral energy metabolism and infarction following transient focal hypoxia-ischaemia in the developing brain

Phosphorus magnetic resonance spectroscopy (31P MRS) was used to determine whether focal cerebral injury caused by unilateral carotid artery occlusion and graded hypoxia in developing rats led to a delayed impairment of cerebral energy metabolism and whether the impairment was related to the magnitude of cerebral infarction. Forty-two 14-day-old Wistar rats were subjected to right carotid artery ligation, followed by 8% oxygen for 90 min. Using a 7T MRS system. 31P brain spectra were collected during the period from before until 48 h after hypoxia-ischaemia. Twenty-eight control animals were studied similarly. In controls, the ratio of the concentration of phosphocreatine ([PCr]) to inorganic orthophosphate ([Pi]) was 1.75 (SD 0.34) and nucleotide triphosphate (NTP) to total exchangeable phosphate pool (EPP) was 0.20 (SD 0.04): both remained constant. In animals subjected to hypoxia-ischaemia, [PCr] to [Pi] and [NTP] to [EPP] were lower in the 0- to 3-h period immediately following the insult: 0.87 (0.48) and 0.13 (0.04), respectively. Values then returned to baseline level, but subsequently declined again: [PCr] to [Pi] at -0.02 h-1 (P < 0.0001). [PCr] to [Pi] attained a minimum of 1.00 (0.33) and [NTP] to [EPP] a minimum of 0.14 (0.05) at 30-40 h. Both ratios returned towards baseline between 40 and 48 h. The late declines in high-energy phosphates were not associated with a fall in pHi. There was a significant relation between the extent of the delayed impairment of energy metabolism and the magnitude of the cerebral infarction (P < 0.001). Transient focal hypoxia-ischaemia in the 14-day-old rat thus leads to a biphasic disruption of cerebral energy metabolism, with a period of recovery after the insult being followed by a secondary impairment some hours later.

The role of the cyclic AMP-responsive element binding protein (CREB) in hypoxic-ischemic brain damage and repair

The cyclic AMP-responsive element binding protein (CREB) is a basally expressed, post-translationally activated transcription factor that has been implicated in the trans-activation of a number of genes in response to cAMP and calcium signals. A unilateral hypoxic-ischemic (HI) injury in the 21 day old rat was used to examine a potential role for CREB (phosphorylated and unphosphorylated) in neuronal programmed cell death or cell survival. The selectively vulnerable CAI pyramidal cells, which undergo delayed neuronal death following mild HI, show a loss of CREB and phosphorylated CREB (pCREB) immunoreactivity on the injured side 48 and 72 h following HI. In contrast the resistant dentate granule cells and cortical cells produce a bimodal increase in pCREB immunoreactivity, peaking 6 and 48 h following HI. The fact that cells surviving the HI insult are showing increased activation of CREB suggests that this protein might be involved in the process of neuroprotection.

Induction of clusterin in the immature brain following a hypoxic-ischemic injury

A unilateral hypoxic-ischemic (HI) insult in the 21 day old rat has been used to assess the role of clusterin in nerve cell death. Both clusterin mRNA and protein levels were measured at various time points after moderate (15 min) and severe (60 min) HI insult using in situ hybridisation and immunocytochemistry respectively. The severe HI insult lead primarily to necrotic neuronal death and showed very little if any clusterin mRNA and protein induction on the ligated side of the brain. However, following the moderate HI insult there was a dramatic time-dependent accumulation of clusterin protein in neurons of the CA1-CA2 pyramidal cell layers in the hippocampus and cortical layers 3-5, regions undergoing delayed neuronal death. Clusterin mRNA expression, in contrast to neuronal protein accumulation, appeared to be glial in origin (probably astrocytes) with increases in mRNA in and around the hippocampal fissure and only a weak signal over the CA1-CA2 pyramidal cell layer. These results support the hypothesis that the clusterin protein is synthesised in the astrocytes, secreted and then taken up by dying neurons. Clusterin immunoreactivity and in situ DNA end-labelling performed on the same sections revealed that clusterin was accumulating in neurons destined to die by programmed cell death. However the relative time-courses of DNA fragmentation and clusterin immunoreactivity suggest that clusterin production was a result of the selective delayed neuronal death rather than being involved in the biochemical cascade of events that cause it.

The effect of prolonged modification of cerebral temperature on outcome after hypoxic-ischemic brain injury in the infant rat

Hypoxic-ischemic injuries can evolve over several days, and recent studies suggest that further neuronal death may occur 6 to 72 h later. Because cerebral temperature is an important determinant of outcome during the primary injury, we investigated the effect of temperature, on outcome, during the later phases of injury. Hypoxic-ischemic injury was induced in 21-d-old rats by unilateral ligation of the right carotid artery followed by exposure to 15 min of hypoxia of 8% O2 at 34 degrees C. Cerebral temperature changes were induced by modifying environmental temperature. The rats were divided into four treatment groups: group 1 (n = 15) remained at 34 degrees C for 72 h; group 2 (n = 14) were kept at 34 degrees C for 6 h and then at 22 degrees C for the remaining 66 h; group 3 (n = 17) remained at 22 degrees C for 6 h and 34 degrees C for the next 66 h; group 4 (n = 16) remained at 22 degrees C for 72 h. Rats kept at 22 or 34 degrees C had cortical temperatures of 35.5 +/- 0.1 degrees C and 37.9 +/- 0.2 degrees C, respectively. Histologic outcome was assessed 72 h after hypoxia. The area of cortical infarction was reduced in group 4 compared with groups 1-3 (p < or = 0.05). Striatal damage was reduced in group 4 (p = 0.05). Hippocampal neuronal loss was not significantly altered. In a subsequent study the area of cortical infarction was 12.1 +/- 3 mm2 in group 1 (n = 11) compared with 3.4 +/- 1.5 mm2 group 4 treated rats (n = 10) 21 d after the injury (p < 0.01). Thus hypothermia spanning both the first 6 h and from 6 to 72 h after injury was needed to improve outcome. Conversely exposure to the thermoneutral environment exacerbated the injury. These observations suggest that prolonged moderate cerebral hypothermia can be used to suppress the cytotoxic processes that occur after hypoxic-ischemic injury.

Mechanisms of delayed cell death following hypoxic-ischemic injury in the immature rat: evidence for apoptosis during selective neuronal loss

The mechanisms leading to delayed cell death following hypoxic-ischemic injury in the developing brain are unclear. We examined the possible roles of apoptosis and microglial activation in the 21-day-old rat brain following either mild (15 min) or severe (60 min) unilateral hypoxic-ischemic injury. The temporal and spatial patterns of DNA degradation were assessed using gel-electrophoresis and in-situ DNA end-labelling. Microglial activation, mitochondrial failure and cell death were examined using lectin histochemistry, 2,3,5,triphenyl-H-tetrazolium chloride (TTC) staining and acid fuchsin staining, respectively. Selective neuronal death produced by the 15 min injury was associated with the development of apoptotic morphology, DNA laddering and acidophilia from 3 days post-hypoxia. The 60 min injury accelerated this process with some cells showing signs of DNA degradation at 10 h post-hypoxia. However, in the cortex, which developed infarction after the 60 min injury, a different pattern of cell loss occurred. The DNA and mitochondria remained intact, and cells basophilic, until after 10 h post-hypoxia, then widespread necrosis developed by 24 hr. In contrast to regions of selective neuronal loss, DNA degradation was initially random (at 24 hr), with 180bp DNA ladders not detected until 3 days post-hypoxia. There was no morphological evidence of apoptosis. Microglial activation coincided with the onset of DNA degradation in regions of selective neuronal loss but not infarction, suggesting a possible role in selective neuronal death. The results suggest that cortical infarction, which was delayed for at least 10 h, was necrotic, and occurred independently of microglial activation and apoptosis. In contrast, selective neuronal death was apoptotic.