86. PTEN/Akt/FOXO3a pathway in neuronal apoptosis in the devoloping rat brain after hypoxia-ischemia
D. Li, Y. Qu and D. Mu
Department of Pediatrics, Westchina Second University Hospital, Sichuan University, Chengdu, China
Background: The pro-apoptotic function of phosphatase PTEN has been linked to its capacity to antagonize the phosphatidylinositol 3-kinase (PI3K)/Akt signaling pathway. However, the roles and the underlying mechanisms of PTEN in neuronal apoptosis after hypoxia-ischemia (HI) are not clear. Previous studies have revealed that Foxhead transcription factor (FOXO3a), a substrate of Akt, is a critical effector of PTEN-mediated tumor suppressor.
Objective: To test whether PTEN/Akt/FOXO3a pathway is involved in neuronal apoptosis in developing rat brain after HI.
Methods: Postnatal day 10 rats were subjected to HI by ligating common carotid artery followed by hypoxia. Immunohistochemistry and Western blot analysis were used to detect the expression and phosphoralation of PTEN, Akt, as well as FOXO3a and its target gene Bim, a pro-apoptotic member of Bcl-2 family.
Results: We found that dephosphorylation of PTEN, Akt, FOXO3a were accompanied by FOXO3a translocation into the nucleus at 0.5 h after HI. Simultaneously, Bim was significantly induced at 0.5 h, peaked at 2 h, preceding the neuronal apoptosis which occurred after 4 h following HI. Furthermore, we found that pretreatment of rats with bpv, a potential inhibitor for PTEN, significantly increased the phosphorylation of PTEN, Akt, and FOXO3a, which leads to a decrease of FOXO3a translocation into nucleus and a decrease of Bim expression after HI. The downregulation of Bim caused by PTEN inhibition decreased cellular apoptosis in the developing rat brain.
Conclusions: Our findings suggest that PTEN/Akt/FOXO3a pathway is involved in neuronal apoptosis in neonatal rat brain after HI. Agents targeting PTEN may offer a promise to rescue neurons from HI brain damage.
87. Unilateral cerebral blood flow perturbations induce cellular responses in mirror in P7 rat brain
C. Charriaut-Marlangue1, S. Villapol1, S. Fau1, P. Bonnin2 and S. Renolleau1
1NPA—Case 14, UMR-CNRS 7102; 2Hôpital Lariboisière, Physiologie—Explorations Fonctionnelles, Paris, France
Objectives: Very few experimental studies have addressed the vascular and haemodynamic responses to arterial occlusion in the developing brain in animal models of brain injury and no studies have been published on mechanisms underlying cerebral blood flow (CBF) perturbations in the immature brain. The aim of this study was to explore the relationship between cerebral haemodynamic changes in P7 rat brain and molecular pathways implicated in cell survival and/or demise.
Methods: Wistar 7 day-old rats of both sexes were anaesthetized (Chloral hydrate 350 mg/kg) and subjected to transient (50 mins) left common carotid artery (CCAo) using a vascular clip. Rats were subjected to ultrasound measurements using an echocardiograph (Vivid 7, GE Medical Systems ultrasound®, Horten, Norway) equipped with a 12-MHz linear transducer as previously reported.1 Doppler spectral recordings in the main arteries were evaluated:
before surgery,
during ischemia and
after reperfusion.
Rats were killed at 48 h post-injury. Caspase-3 (C3) cleavage (p17 protein) was monitored either by immunohistochemistry on cryostat sections or by western blotting. DNA fragmentation was determined using the TUNEL assay and cell death by Fluoro Jade B.
Results: Unilateral transient CCAo (uni-tCCAo) did not induce cell death (no TUNEL and Fluoro Jade B staining) but induced cleavage of C3 in both ipsilateral (IL) and contralateral (CL) cortical layers II–III and VI. Using double immunofluorescence and confocal analysis we determined that cleaved C3 was mainly located in neurons (NeuN-positive) and in immature cells (nestin-positive). Cleaved C3 was expressed by GABAergic neurons (layer III). Blood flow monitoring indicated that during ischemia left ICA (internal carotid artery) was no injected. In contrast, right (CL) ICA and basilar trunk (BT) mean blood flow velocities (mBFVs) significantly increased (by 40%, P<0.01% and 39%, P<0.05, respectively) to supply the arterial terminal branches of the left ICA through the anterior and posterior communicating arteries. After release of the left uni-tCCAo, left ICA was re-injected and left ACA was anterogradelly injected but these two arteries exhibited reduced mBFVs (by 50% and 45%, P<0.05, respectively). However, mBFV was not significantly modified in the CL middle cerebral artery during ischemia and after reperfusion. A negative significant correlation between the percentage of immediate CCA reflow and the number of cells positive for cleaved C3 was found in both sides. Upstream caspase-2 cleavage and cytochrome c release from mitochondria were detected in both IL and CL cortex tissues at 48 h post-injury.
Conclusions: This study represents the first demonstration that an ipsilateral (IL) hemodynamic stress (50 mins left CCA occlusion), not sufficient to induce cell death, produces a bilateral and symmetric cellular response in the MCA territory giving images in mirror in the developing brain. Data suggest that IL energetic deficit only produces upstream apoptotic markers in GABA interneurons (in layer II-III) and glutamatergic neurons (in layer VI), which can transmute signals to the CL side either by transcallosal fibers and/or directly by their long distance axons.
97. Abnormal responses of thalamocortical neurons after asphyxial cardiac arrest in developing rats
M. Shoykhet1,2, D.J. Simons3, P.M. Kochanek1,2 and R.S. Clark1,2
1Critical Care, University of Pittsburgh School of Medicine; 2Safar Center for Resuscitation Research, University of Pittsburgh; 3Neurobiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
Background: Children who survive cardiac arrest (CA) often suffer from devastating neurologic consequences. Yet, how CA disrupts behaviorally-relevant neuronal circuits during development remains unknown. The rat somatosensory system is well-suited for defining in vivo neuronal function after CA because it mediates sophisticated sensorimotor behaviors and because its thalamocortical circuitry is well characterized. Neurons in the thalamic ventrobasal (VB) nucleus receive sensory input, process it, and relay it to the cerebral cortex. Neurons in the thalamic reticular (RT) nucleus provide essential inhibitory regulation of VB neurons. RT neurons are thought to be particularly vulnerable to hypoxic-ischemic insult. However, the extent of this vulnerability and its impact on functional organization of developing thalamic circuitry are unknown.
Objective: To characterize responses of VB neurons to tactile stimuli after asphyxial CA in developing rats.
Design/methods: Postnatal day 17 rats underwent either sham insult or asphyxial CA followed by clinically-realistic resuscitation. Two to 3 days post-CA, we utilized extracellular microelectrodes to record responses of single VB neurons to controlled whisker deflections in different directions. Spontaneous and evoked firing rates, as well as firing evoked by movements of different whiskers in multiple directions were compared in sham and CA rats. After recordings, the rats were anesthetized, and the brains were fixed for immunohistochemical analyses.
Results: We recorded responses of 21 and 28 VB neurons in 3 sham and 3 CA rats, respectively. VB neurons in CA animals, like those in sham rats, have a clearly-identifiable response to one or a few neighboring whiskers. However, after CA, VB neurons display abnormally high spontaneous and whisker-evoked firing rates. VB neurons are also somewhat less selective for whisker movement in different directions in CA versus sham rats. In addition, responses occur at longer latency post-CA, suggesting abnormalities in response timing.
Conclusions: Results suggest that thalamic input to the cerebral cortex is abnormal in survivors of asphyxial CA. Our preliminary findings are consistent with injury to inhibitory RT neurons and suggest that the thalamus is a potentially important therapeutic target after asphyxial CA. Further characterization of post-CA thalamic neuropathology and alteration in thalamocortical circuit function are ongoing.
104. Color-coded pulsed doppler ultrasound imaging predicts ischemic cerebral lesion in rat pups
C. Charriaut-Marlangue1, N. Deroide2, S. Renolleau1, P.-L. Leger1, S. Fau1 and P. Bonnin2
1NPA—Case 14, UMR-CNRS 7102; 2Hôpital Lariboisière, Physiologie—Explorations Fonctionnelles, Paris, France
Background and aims: Models of ischemic brain lesion in neonatal rats displayed heterogeneous lesion volumes. Moreover, several rat pups could present an absence of brain lesion, whenever ischemic procedure is well applied. Because of those variations, neuroprotection studies need numerous animals to demonstrate any significant inter-group difference. Color-coded duplex ultrasound give access to measurement of blood flow velocities in cerebral arteries in small animals. We tested whether absence of cerebral lesion could depend on intra cerebral hemodynamic adaptations during and/or after procedure of cerebral ischemia-reperfusion.
Methods: Blood flow velocities of cerebral arteries were measured by ultrasound imaging (Bonnin et al, 2008) in 39 anaesthetized (chloral hydrate 350 mg/kg) Wistar 7 day-old rats:
before,
during permanent left middle cerebral artery and transient left common carotid artery (50 mins) occlusions (Renolleau et al, 1998) and
after release of occlusion of the left common carotid artery.
Results: Forty-eight % of rats did not present an ischemic brain lesion. In those rats, ultrasound imaging evidenced:
an early rise of the blood flow velocities in large upstream arteries and
increased blood flow velocities in Willis's circle arteries giving anastomoses to the left MCA territory.
Conclusions: Ultrasound imaging can predict the absence of ischemic cerebral lesion despite arterial occlusions. Homogeneous groups of rat pups all presenting ischemic brain lesions can be built, making, thus, preclinical neuroprotection studies easier, less time-consuming and more specific and cost-effective by lowering the intra-group variations and the required number of animals.
145. Age-dependent effects of gradual cerebral perfusion pressure (CPP) decrease and temporary CPP breakdown on neurochemical response in swine
R. Bauer1, B. Walter2, K. Aisenpreis1, J. Soukup3, M. Eiselt4 and H. Fritz5
1Institute of Molecular Cell Biology, Friedrich Schiller University, Jena; 2Department of Neurosurgery, Zentralklinik Bad Berka, Bad Berka; 3Department of Anesthesiology and Intensive Care Medicine, Martin Luther University Halle-Wittenberg, Halle/Saale; 4Institute of Medical Statistics, Computer Sciences and Documentation, Friedrich Schiller University, Jena; 5Department of Anesthesiology and Intensive Care Medicine, Matha-Maria Hospital, Halle/Saale, Germany
Objectives: Compromised CPP appears as a main pathogenetic factor for disturbed brain functioning and resulting structural brain damage during ontogenesis. However, there is still a lack of knowledge on the age-dependent relation between the level of CPP reduction, which induces a critically reduced brain perfusion leading to excessive transmitter release and disturbed brain energy metabolism prone to initiate brain tissue damage cascades. Therefore, in newborn and juvenile pigs, we gradually reduced CPP up to a complete temporary breakdown and estimated the effects on brain oxidative metabolism as well as responses on extracellular glutamate (Glu), dopamine (DA), lactate (Lac) and lactate/pyruvate ratio (Lac/Pyr).
Methods: Experiments were performed in 8 newborn (PD13) and 11 juvenile pigs (PD51), anesthetized with 1.3% isoflurane in N2O/O2 and artificially ventilated. CBF was measured by colored microspheres, CMRO2 and CMRGlucose were calculated. Intracranial pressure (ICP) was measured subcortically and two microdialysis probes were implanted bilaterally into the striatum. Extracellular DA, Glu, Lac and Lac/Pyr were estimated by HPLC with electrochemical detection and by enzymic fluorometric assays. For CPP manipulation, cisterna magna was punctured by a lumbar puncture needle and fixed in place for elective CSF (cerebrospinal fluid) infusion/withdrawal to control ICP. Furthermore, a cerclage of a plastic-coated wire was performed around the trunk of the pulmonary artery in order to apropriately adjust the vessel diameter for perfusion control resulting in ABP stabilization (in order to prevent Cushing response). Artificial CSF infusion was adjusted to induce consecutively a stepwise CPP decrease at four levels, beginning at 50 mm Hg (CPP-50), followed by 40 mm Hg (CPP-40), 30 mm Hg (CPP-30), and zero mm Hg (CPP-0, ICP was always ∼10 to 20 mm Hg above ABP) for every 15 mins. The recovery period was observed for 180 mins.
Results: Whereas juvenile pigs showed a corresponding gradual decrease in CBF between stages CPP-50 and CPP-30, a significant CBF reduction occurred in newborn piglets not until stage CPP-30 (P<0.05). At this stage of CPP reduction an alteration of brain oxidative metabolism occurred solely in juvenile pigs (P<0.05). This was accompanied by DA and Glu elevation (P<0.05). However, already at CPP-50, Lac and Lac/Pyr were elevated (P<0.05). In contrast, newborn piglets showed mild elevation in Lac and Lac/Pyr not until stage CPP-30, whereas all other microdialysis parameters changed during the period of complete global brain ischemia at CPP-0. Complete global brain ischemia was confirmed by nearly complete suppression of ECoG activity. At this stage striatal DA and Glu reached the strongest elevation in juvenile pigs. Similar alterations were seen in newborn piglets, but the respective changes were markedly blunted (P<0.05).
Conclusions: Lower limit of CBF autoregulation determines modifications in neurochemical parameters, clearly before CMRO2 and CMRGlucose are reduced. Early indicators for mild to moderate hypoperfusion are evidently Lac, Lac/Pyr and DA elevation, but Glu appears to be an indicator for tissue ischemia. The immature brain generates a markedly diminished potential for excitotoxicity suggesting a substantial reason for blunted cerebral damage in consequence of gradual CPP impairment.
154. Administration of fish oil during perinatal periods greatly relieve the behavioral deficits in neonatal hypoxia-ischemia
Y. Gao1,2, W. Zhang1,2, H. Gao1, W. Yang1, H. Shen1, G. Cao1,2 and J. Chen1,2
1State Key Laboratory of Medical Neurobiology, Fudan University, Shanghai, China; 2Department of Neurology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
Background: The development of the rodent models of neonatal hypoxia-ischemia brain injury is similar to what is observed in the human neonates suffering from hypoxia-ischemia encephalopathy. In spite of increased knowledge of its pathophysiology, we still lack effective strategy for neuroprotection. OMEGA-3 play an important role in many aspect of physical function such as synaptic transmission, cognitive development and memory-related learning. Fish oil is considered to be the major source of OMEGA-3, especially content high percentage of Eicosapentaenic acid (EPA) and Decosahexaenic acid (DHA). DHA is a critical component of brain membrane phospholipids and is necessary for the development of nervous system. EPA could inhibit the conversion of arachdonic acid (AA) into and prostaglandins, which is the main source of inflammation after brain injury, and is a potential antioxidant against ischemia brain injury. In this study, we determined the changes of OMEGA-3 in brain tissue after neonatal hypoxia-ischemia brain injury and investigated the neuroprotective effect of fish oil on the behavioral recovery after hypoxia-ischemia.
Methods: After being pregnant, rats were divided into two groups, one is given regular diet for rodent (RD group) and the other is supplied special diet containing 5% fish oil (FOD group), then both of them were given regular diets 14 days after gestation. Neonatal of 7 days old was suffered from hypoxia-ischemia brain injury, and Lipid analysis was performed respectively on 1 day or 7 days after this injury. Behavioral performance of gait walking, righting reflex, grid walking and water maze were also recorded after damage. All data are expressed as mean±s.d., and statistic analysis was performed using SPSS tests, with P<0.05 considered statistically significant.
Results: We found that, with the increase of AA, hypoxia-ischemia brain injury greatly reduced the concentration of DHA and EPA in brain tissue. Although only hypoxia could decrease the ratio of DHA and EPA to AA (from the normal ratio of 2.68±0.26 to 2.12±0.11, P<0.001) 24 h after hypoxia, but it raise to the normal level 7 days after H-I. On the other hand, Hypoxia-Ischemia cause long last decrease of this ratio (from 2.44±0.26 to 1.95±0.03, P<0.001) fish oil treatment could maintain the high level of this ratio even after the hypoxia-ischemia (RD group: from the normal 2.68±0.26 to 2.04±0.32, 24 h after H-I, P<0.01; FOD group: from the normal 3.05±0.09 to 3.15±0.10, 24 h after the damage, P<0.05). The concentration of EPA in ipsilateral is extremely higher compared with sham in fish oil diet group 7 days after the injury (sham is 0.138±0.01%, H-I is 0.176±0.009% respectively, P<0.05). Furthermore, fish oil apparently relieve the behavioral deficits in gait walking and righting reflex (P<0.001), and improve the long term recovery in grid walking and water maze even 5 weeks after brain damage (P<0.05).
Conclusions: These results suggest that supplementary of fish oil could ameliorate the brain damage and is beneficial to the behavioral recovery from hypoxia-ischemia brain injury.
200. Effects of Na+/H+ exchanger isoform 1 inhibitor cariporide on CA1 pyramidal neuronal death in mouse model of hypoxic ischemic encephalopathy
P. Cengiz1, K. Uluc2, N. Kleman2, P. Ferrazzano1 and D. Sun2
1Department of Pediatrics, University of Wisconsin; 2Department of Neurological Surgery, University of Wisconsin, Madison, Wisconsin, USA
Background: Hypoxic ischemic encephalopathy (HIE) is an important cause of mortality, morbidity in neonates. Goal of this project is to investigate role of plasma membrane Na+/H+ exchanger isoform 1 (NHE-1) in neonatal HIE. NHE-1 is essential in regulation of pHi. Its activity is stimulated upon acidosis after ischemia. We previously found that inhibition of NHE-1 is neuroprotective following focal cerebral ischemia in adult animals (Luo et al, 2005). However, it is unknown whether NHE-1 plays a role in immature brain damage following hypoxia ischemia (HI). We hypothesized that overstimulation of NHE-1 occurs in HI and inhibition of NHE-1 with cariporide (HOE 642) is neuroprotective.
Methods: Left common carotid arteries of P9 mice (C57/Blk6) were ligated under isoflurane anesthesia and 0.21% FiO2. Animals recovered for 2 hours (h), exposed to 8% O2 (balanced nitrogen) for 55 mins (min) at 37°C and were grouped into 4 groups. Group I and III received saline injection intraperitoneally (i.p) 3 times as vehicle controls. Group II received HOE 642 (0.5 mg/kg i.p.) 10 mins, at 24 and 48 h after HI. Group IV received HOE 642 (0.5 mg/kg i.p.) 5 mins prior to initiation of HI, and at 24 and 48 h after HI. At 72 h after HI, the brains were transcardially fixed with 4% paraformaldehyde. Brain sections (70 μm) were prepared and stained with Fluoro-Jade. Number of Fluoro-Jade stained cells in the CA1 regions of hippocampus were compared between the vehicle control and HOE 642-treated animals. Neuronal damage was quantified using a scale (score 1: 0 to 200, 2: 201 to 500, 3: 501 to 1000, 4:>1000 positively stained cells). Four slices for each brain were analyzed. Number of Fluoro-Jade positive cells per brain was averaged.
Purpose: To investigate whether pharmacological inhibition of NHE1 reduces hippocampal damage in neonatal mice and to investigate dose and administration interval of HOE 642 for neuroprotection following HI.
Results:Figure 1 shows, CA1 pyramidal neurons in vehicle control mice were positively stained with Fluoro-Jade (left panel, arrow) at 72 h following 55 mins HI. In contrast, the HOE-treated mouse had few stained cells in CA1 region (right panel, *). ∼30% reduction of Fluoro-Jade-positive cells was seen in HOE-treated brains (groups II and IV) (Table 1).
Conclusions: This preliminary study demonstrated that inhibition of NHE-1 with HOE 642 following HI was neuroprotective in P9 mice suggesting that NHE-1 activity may contribute to delayed CA1 pyramidal neuronal cell death after neonatal HI.
Neurological injury scores of the groups
Score (mean±s.d.)
Group I
2.7±1.2
Group II
1.9±1.1
Group III
2.0±1.1
Group IV
1.9±1.1
238. Superfused neonatal rat cerebrocortical slices during oxygen-glucose deprivation: 1H/31P NMR metabolic characteristics of bioenergetic protection by mild hypothermia
J. Liu1, M.J.S. Kelly2 and L. Litt1
1Department of Anesthesia and Perioperative Care; 2Department of Pharmaceutical Chemistry, University of California San Francisco, San Francisco, California, USA
Aims: Large clinical trials of asphyxiated human neonates have found that mild therapeutic (post-insult) hypothermia substantially improves neurological outcomes. Using an ex vivo rodent brain slice model and high resolution (14T) NMR determinations of Perchloric Acid (PCA) extracted metabolite concentrations, we asked if NMR metabolic profiles after oxygen-glucose deprivation (OGD), during and after mild hypothermia, could characterize bioenergetic impairment and recovery, accurately differentiating good from bad outcomes, thereby encouraging additional data taking for multivariate metabolomic analyses.
Methods: Experiments began with 20 superfused, respiring, 350 μ cerebrocortical slices obtained from the same litter of 7-day old rat pups, using the IACUC-approved protocol in earlier studies.1 The superfusate, or oxyACSF, was a modified Krebs balanced salt solution and 10 mmol/L glucose.1 Bicarbonate buffering and oxygenation of oxyACSF was from continuous bubbling with (95% O2)/(5% CO2) gas mixture, resulting in constant PCO2 (40 mm Hg), PO2 (600 to 650 mm Hg), pH (7.4), and temperature (37°C for normothermia; 32°C for 3 h of hypothermia followed by 60 mins slow rewarming.) Thirty minute OGD experiments began after 3-h oxygenation with no change in the 10 to 15 mL/mins flow, when oxyACSF was switched to ACSF with no glucose and a PO2 <8 mm Hg (Ocean Optics oximeter) from 3 h of bubbling 100% N2. In one group hypothermia began just before OGD, in another it began after 15 mins of OGD. Five slices were removed: before starting OGD; at the end of OGD before restoring oxyACSF; at the end of hypothermia, or at the corresponding time during normothermia; and after 3 h more of normothermic recovery. Twenty-eight PCA metabolites were quantified with 1H/31P NMR analysis. Histology and immunohistology was used to examine slices removed at different times in the protocol.
Results: In slices where 3 h hypothermia began just before OGD, concentrations at the end of the experiment (after slow rewarming and 3 h normothermia) of 8 metabolites were significantly (P<0.05) greater in hypothermia-treated slices, with the following (hypothermia treated group)/(normothermia treated group) ratios: alanine (1.6), acetate (1.32), NAA (1.8), GABA (1.65), glutamate (1.72), glutamine (2.14), and ATP (1.33). Because we had N = 4, it is likely that more data will provide additional significant concentration differences. In additional experiments where the onset of 3 h hypothermia began 15 mins after the start of OGD, final concentration ratios (hypothermia/normothermia) for the 8 metabolites were still significantly greater than 1. As expected, slow rewarming always caused significant changes in metabolite concentrations.
Conclusions: Finding metabolic/histologic protection when mild hypothermia starts both before and after OGD validates use of the neonatal brain slice model for mechanistic studies. The data are sufficiently accurate to proceed to metabolomic investigations. We are optimistic that metabolomic multivariate analyses will ultimately be able to distinguish and predict good and bad outcomes.
266. The role of activated microglia/macrophages in neonatal focal stroke
Z. Vexler1, J. Faustino1, A. Klibanov2, C. Jonhson1, N. Derugin3 and M. Wendland4
1Neurology, University California San Francisco, San Francisco, California; 2University of Virginia, Charlottesville, Virginia; 3Neurosurgery; 4Radiology, University California San Francisco, San Francisco, California, USA
Objectives: Studies in adult stroke models have demonstrated that pharmacologic strategies that limit macrophage accumulation/microglial activation are neuroprotective, suggesting that microglial cells contribute to injury induced by cerebral ischemia. Recent studies, however, showed that selective ablation of proliferating microglial cells exacerbates ischemic injury in the adult brain and that microglial cells can support neurogenesis, indicating beneficial properties of these cells. In the neonatal brain, macrophage accumulation is substantial both after hypoxia-ischemia and focal ischemic stroke but the relative contribution of activated microglia and invading macrophages in injury is unknown. We have previously reported that early after transient focal cerebral ischemia in neonatal rats macrophages are comprised of activated microglia rather than invaded macrophages (Denker et al, J Neurochem 2007). The aim of this study was to determine if depletion of microglial cells affects injury acutely after neonatal focal stroke.
Methods: Postnatal day 5 (P5) rats were injected with liposome-encapsulated clodronate or empty liposomes (3ul, intra-cortical). At P7, rats were subjected to a 3 h middle cerebral artery occlusion (MCAO) and underwent diffusion-weighted MRI during occlusion for evidence of injury. Injured rats were sacrificed 24 h post-reperfusion for measurements of injury volume (determined in 6 consecutive Nissl-stained coronal sections), densities of microglial cells in adjacent sections (Iba1 immunohistochemistry), and numbers of neurons with cleaved caspase-3. Cytokine concentrations (8-plex, IL1a, IL1b, TNF-a, IL-6, IL-18, CINC-1, MCP-1 and MIP-1) and degradation of a structural protein spectrin by caspase-3 and calpain (Westernn blot analysis) were measured in injured and contralateral cortex in separate groups of rats.
Results: Depletion of microglia by clodronate was confirmed in injured rats by lack of Iba1 staining whereas the expected pattern of ramified microglial cells in contralateral hemisphere and ameboid cells in injured tissue (Dingman et al, J Neurochem 2007) was observed in vehicle-treated rats. Injury volumes were not significantly different between clodronate-treated and vehicle-treated groups (51.1±7.4% and 54.8±8.3%, respectively, n = 5 to 6 per group). Spectrin cleavage mediated independently by calpains and caspase-3 was profoundly increased in injured tissue in both groups, as was evident from Western Blot analysis (11 to 15 and 20 to 25 fold increase for calpain- and caspase-3 cleaved spectrin, respectively), but spectrin degradation was not significantly affected but clodronate treatment. Microglial depletion affected cytokine concentrations in contralateral hemisphere, IL-6 levels in particular. Surprisingly, in ischemic-reperfused tissue, the levels of IL1b, TNF-a, IL-6, CINC-1 and MCP-1 which were elevated by MCAO, remained unaffected and only the levels of MIP-1a were significantly reduced following microglial depletion.
Summary and conclusions: Our findings demonstrate that in neonatal rats depletion of microglial cells does not reduce injury size after focal ischemia-reperfusion, at least acutely. We are currently exploring whether depletion of microglial cells results in reduction of phagocytosis of apoptotic neurons and enhances injury over time.
NIH NS44025, AHA GIA 0855235F.
305. Routine DWI-MRI as a bias-killer and a rat pup-saver in neonatal neuroprotection preclinical trials
S. Fau1,2, C. Po3, P. Meric3 and C. Marlangue1
1NPA, UPMC CNRS UMR 7102; 2NICU/PICU, Hopital d'Enfants A. Trousseau/APHP, Paris; 3ICSN, CNRS, Gif-sur-Yvette, France
Objectives: Lack of internal validity in preclinical trials may lead to disappointing failures in clinical trials. Moreover biases may increase preexisting variability of animal models and consequently increase requested sample sizes.
In previous work we highlighted that our model of transient neonatal focal ischemia suffers from a poor reproductibility because of:
inconstant initial ischemia,
cases of poor reperfusion,
variability of the concerned arterial territory size.
This leads to high variance of the infarct size and finally increases required sample size. We make the hypothesis that post-reperfusion DWI-MRI allows to:
detect absence of initial ischemia,
detect poor reperfusion,
delineate the initial ischemic territory.
Our objective is to prove that routine DWI-MRI in our neuroprotection studies allows to remove biases and use less animals.
Methods: A neuroprotection study tested the effectiveness of the pananticaspase Q-VD-OPh to reduce the H48 infarct size in our model (1 mg/kg IP directly after reperfusion). This model, performed in P7 Wistar rat pups, associates a permanent MCA occlusion with a transient CCA occlusion (50 mins). Using previous data, requested sample size calculation leads to n = 24 per group (α = 0.05, 1−β = 0.80, expected effect: 40% decrease of infarct size, bilateral test). Using a 7.4T, we performed ADC map and MR-angiography in each animal 10 and 120 mins after reperfusion. We evaluate postischemic edema on ADC maps (VADCM10 and VADCM120) by counting in ipsilateral hemisphere, pixels which ADC was < mean-2s.d. of the controlateral hemisphere ADCs. We named ‘artefacts’ the animals presenting an absence of postischemic edema and/or a poor reperfusion at M10 post-reperfusion. Classical histology measurement of the infarct size (VHISTO) was carried out at H48. We created two composite measurements using the ratios of final infarct size and postischemic edema sizes, respectively VHISTO/VADCM10 and VHISTO/VADCM120. We compared 4 various statistical analysis using the various measurement and including or not the ‘artefacts’, respectively: VHISTO including ‘artefacts’, VHISTO excluding ‘artefacts’, VHISTO/VADCM10 excluding ‘artefacts’ and VHISTO/VADCM120 excluding ‘artefacts’.
Results: On 48 animals, 2 died, 20 didn't show any initial edema, 4 showed poor reperfusion. None of the four statistical analysis showed effectiveness of the Q-VD-OPh. A posteriori power was extremely low when MRI was not used (0.16). A posteriori power was much higher when we associated use of composite criteria and exclusion of ‘artefacts’. It arose that, a posteriori, requested sample size (calculated with above conditions) decrease from 94 down to 9 per group.
Conclusions: In neuroprotection studies using our model, routine DWI-MRI allows on the one hand to eliminate biases like absence of initial ischemia or poor reperfusion, and on the other hand to dramatically reduce the requested sample sizes.
Main results
Measurement/artefact
Vhisto/included
Vhisto/excluded
Vhisto/vadcM10/excluded
Vhisto/VadcM120/excluded
Vehicle (moy±s.d.)
7.2±7.3
11±5.1
54.5±17.3
62.5±14.8
Q-VD-OPh (moy±s.d.)
8.0±7.7
10.7±6.9
56.9±26.1
59.3±22
p (t-test)
0.71
0.91
0.80
0.69
a posteriori power
0.16
0.40
0.67
0.88
n per group in a future study
94
30
15
9
323. MRI and histological changes corresponding to descending corticospinal tract degeneration following neonatal hypoxic ischemic brain injury and associated behavioral outcome
S. Lama1, M. Qiao2, A. Ng1, D. Barua1, D. Kirk1, T. Foniok2, A. Kirton3,4 and U. Tuor1,2
1Medical Science, University of Calgary; 2MR Technology, Institute for Biodiagnostics (IBD-West), National Research Council; 3Pediatrics and Clinical Neurosciences, University of Calgary; 4Pediatrics (Division of Neurology), Alberta Children's Hospital, Calgary, AB, Canada
Background and aims: Recently, clinical magnetic resonance imaging (MRI) studies have demonstrated acute diffusion weighted imaging (DWI) changes in the descending corticospinal tract (DCST), which includes motor fibres of the posterior limb of the internal capsule, cerebral peduncle, basis pontis and medullary pyramid.1,2 This has been attributed to possible early neuronal injury or Wallerian degeneration but the corresponding tissue correlates are currently not known. Crucial for improved diagnosis is the availability of an animal model with comparable MRI changes and in which the corresponding tissue changes of this pathological process can be determined and translated clinically.
Methods: Seven day old Wistar rat pups (n = 29) were subjected to either sham surgery or unilateral cerebral hypoxia-ischemia produced by occlusion of the right common carotid artery under isoflurane anesthesia followed by exposure to hypoxia.3 At 24 and/or 48 h following the hypoxia-ischemia, animals were anesthetized with isoflurane and T2, Diffusion weighted images (DWI) and Apparent Diffusion Coefficients (ADC) maps of the brain were acquired using a 9.4T Bruker Biospin (Magnex) MR system. Animals were euthanized immediately post MRI or allowed to recover up to 4 weeks post insult. Behavioral outcome was assessed, 4 weeks post injury, using a modified neurological scoring system and cylinder test. Left-right hemispheric differences in T2 relaxation times and ADC were compared using a paired t-test and to sham control values. Brains were processed and stained immunohistochemically with SMI 31 antibody for axonal phosphorylated neurofilaments.
Results: Quantitation of the T2, DWI intensity and ADC changes demonstrated significant left-right differences along the DCST at 24 and 48 h post injury (Figure 1). Differences were most prominent in the cerebral peduncle and internal capsule (Figure 2). SMI 31 staining of neurofilament within axons generally showed loss of staining ipsilaterally. Behavioral scores indicated deficits in the hypoxic ischemic animals in comparison to the sham controls.
Mean T2 or ADC in left or right (ips lateral) regions of the DCST following either at 24h (A) or 48h (B) after a unilateral cerebral hypoxic ischemic insult. (*p<0.001. † p<0.05: Py Tr=Pyramidal tract. cp=Cerebral peduncle. ic=internal capsule, post=posterior, pno=Pontine reticular nuclei oral)
Representative diffusion weighted images at the given times post suregary at medullary (A), pontine (B) and midbrain (C) levels. Arrows show cerebral peduncle and internal capsule of the DCST in B and C respectively.
Conclusions: CNS Wallerian degeneration detected histologically occurs in response to a substantial cortical injury caused by cerebral hypoxia ischemia and this can be detected by MRI in the form of T2, DWI and ADC changes. This is accompanied by an observation of behavioural deficits in the hypoxic ischemic group. Our study provides evidence of a novel animal model to further investigate and help establish axonal degeneration as clinicopathological diagnosis non-invasively. (Supported by Heart and Stroke Foundation, Alberta and AHFMR/HBI Traineeship 2008).
419. Protective effects of PACAP and VIP on ischemia-sensitive vascular reactions of the neonatal brain
F. Bari1, L. Lenti1, A. Zimmermann1, D. Kis1, O. Oláh1, G. Tóth2, D. Busija3 and F. Domoki1
1Department of Physiology; 2Department of Medical Chemistry, Faculty of Medicine, University of Szeged, Szeged, Hungary; 3Department of Physiology and Pharmacology, Wake Forest University Health Sciences, Winston-Salem, North Carolina, USA
Pituitary adenylyl cyclase activating polypeptide (PACAP) and vasoactive intestinal peptide (VIP) are structurally related peptides sharing at least partially the same receptors. Both neuropeptides are potent vasodilators of the cerebral circulation, and also, they are neuroprotective in numerous experimental models. Still, it is already evident that these two peptides are not completely identical in their action and each peptide activates multiple mechanisms. In piglets, PACAP38 dilates pial arterioles via cyclooxygenase (COX)-dependent mechanisms, while PACAP27-related vasodilation is insensitive to COX enzyme blockade.1 The effect of VIP on the newborn pial vasculature has not been characterized yet. Impairment of cerebrovascular reactivity (CR) contributes to ischemia/reperfusion (I/R)-induced neuronal damage. However, neuroprotective mediators such as PACAP and VIP may be released from the neurovascular unit during I/R, which could reduce neuronal injury. In our present work we sought to characterize the VIP-induced vasodilation. Furthermore, we tested if PACAP and VIP preserve CR to endothelium- or neuron-dependent I/R-sensitive dilator responses. Pial arteriolar diameters were determined via the closed cranial window/intravital microscopy technique in anesthetized, ventilated piglets (1 day old, n = 86). First, we tested the vasodilator potency of VIP (10−8 to 10−6 M, topically) on the newborn pial vasculature. Prior to the repeated VIP application vehicle, the non-selective COX inhibitor indomethacin (5 mg/kg) or the selective COX-1 and COX-2 inhibitors SC-560 (1 mg/kg) and NS-398 (1 mg/kg) were administered intravenously. The involvement of nitric oxide synthase (NOS) was also evaluated using the NOS inhibitor N-omega-nitro-l-arginine methyl ester (L-NAME 15 mg/kg iv). In the second set of experiments vascular responses to hypercapnia (5% to 10% CO2 ventilation) or topical N-methyl-D-aspartate (NMDA, 10−4 M) were measured before and after I/R. Prior to I/R, non-vasoactive doses of PACAP27, PACAP38 (10−8 M), VIP (10−9 M), or vehicle were applied onto the cortex. VIP evoked concentration-dependent, repeatable pial arteriolar dilation (15±3 to 69±8%*, mean±s.e.m., n = 8, *P<0.05) in our model. This vascular response was partially COX-1-dependent, since indomethacin and SC-560 abolished the vasodilation at the lowest VIP concentration (15±4% and −1±3%* dilation before and after indomethacin, n = 7); however, arteriolar responses remained intact when applying higher doses. I/R significantly attenuated hypercapnia- and NMDA-induced vasodilations that were both preserved by either PACAP27 or 38. For instance, post-ischemic CR to NMDA (vehicle versus PACAP27, CR expressed as % of response before I/R, n = 6 to 6) was 31±10% versus 94±10%*, and CR to 10% CO2 was 27±8% versus 101±19%* (n = 6 to 6). VIP also protected the CO2-evoked response (88±13%* CR to 10% CO2 after I/R, n = 8), however, failed to preserve the NMDA-induced vasodilation after I/R (35±12% CR after I/R, n = 6). In conclusion, VIP is potent vasodilator in the neonatal cerebral circulation by activating (at least at low concentration) COX-1-dependent pathways. PACAP- and VIP-induced neuroprotection is likely mediated in part by preservation of certain I/R-sensitive cerebrovascular mechanisms via at least partially different, specific ways of action.
442. IL-10 overexpression by non-viral gene therapy after excitotoxic neonatal brain damage
P. Gonzalez1, L. Acarin1, H. Peluffo2, A. Aris3, A. Villaverde3, B. Castellano1 and B. Gonzalez1
1Neuroscience Institute and Department of Cell Biology, Physiology and Immunology, Autonomous University of Barcelona, Barcelona, Spain; 2Pasteur Institute of Montevideo and Department of Histology and Embryology, Faculty of Medicine, UDELAR, Montevideo, Uruguay; 3Institute of Biology and Biotechnology, Autonomous University of Barcelona, Barcelona, Spain
Background and aims: As a result of its strong anti-inflammatory nature, together with the well known involvement of inflammation in determining lesion outcome of central nervous system (CNS) pathologies, interleukin-10 (IL-10) has shown a great neuroprotective potential after adult CNS injuries. However, very little is known about its putative anti-inflammatory and neuroprotective role in the damaged immature CNS, where its specific particularities, suggest a differential scenario for IL-10 actions. Therefore, the aim of the present study was to evaluate the anti-inflammatory and neuroprotective potential of IL-10 overexpression after neonatal brain excitotoxic damage.
Methods: IL-10 overexpression was induced by means of non-viral gene therapy in NMDA-injected postnatal day 9 rat brain. Following several survival times ranging from 12 h until 7 days post-lesion, samples were processed for PCR techniques, ELISA analysis, western blotting, histological techniques and double immunofluorescent labeling.
Purpose: Evaluate changes in lesion volume, neurodegeneration rate, astroglial and microglial activation, neutrophil infiltration, production of pro-inflammatory cytokines and expression inflammatory related enzymes induced by IL-10 overexpression after neonatal excitotoxicity.
Results: IL-10 overexpression increased lesion volume and the density of degenerating neurons at early survival times after the excitotoxic damage. In parallel, IL-10 treated animals displayed an increased density of microglia/macrophages and a reduced astroglial content of intermediate filament proteins like GFAP and vimentin. Moreover, IL-10 overexpression enhanced neutrophil recruitment to the brain parenchyma and production of inflammatory related enzymes such as cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS). Finally, IL-10 overexpression did not induce any variation in interleukin-1β and tumour necrosis factor α production but a slight increase in interleukin-6 content at later survival times.
Conclusions: IL-10 overexpression in the excitotoxically damaged postnatal rat brain is unexpectedly deleterious for lesion outcome and modulates the main associated inflammatory processes.
572. Empirical study of intra-utero fetal cardiac intervention of near-term fetal lamb
Y.-M. Hua1, K.-Y. Zhou1, X.-Q. Shi1, Y.-B. Wang1, Z. Zheng1 and Q. Zhu2
1Pediatric Cardiology Department; 2Department of Ultrasound Cardiography, The Second West China Hospital of SiChuan University, Chengdu, China
Objective: To evaluate the feasibility of analogous intrauterine fetal cardiac intervention in near-term fetal lambs.
Methods: Nine bigeminal pregnant ewes in latter 2nd-trimester and early 3rd-trimester were taken into the study of fetal cardiac intervention. Observated the end-results of after-procedure gestation. Evaluated histopathologic changes of brain, lung and liver in fetal and newborn lambs.
Results: While cardiocentesis the fetal lambs' heart rate decreased transiently and there exited bigeminal premature ventricular contractions in two procedures. Blood glucose and lactic acid were found increaseing significantly after procedure (P<0.05). During procedure fetal electrolyte kept stabilization. 3/9 ewes executed after fetal intervention, the ewes which come to full-term delivery and prematurity were both 3/9. The fetuses after-procedure were delivered through vagina and alive. The surfaces of the fetal chest were without inflammation and scar formation. Histological examination of brains, lungs and livers from both control and test groups demonstrated that all fetal lambs' brains were normal except that only one brain tissue of the test group showed mild pathologic change and all fetal lambs' lungs were normal. Immunohistochemisty of fetal brain showed there exited mild nerve cells apoptosis in three fetal brain with TUNLE. The fetal hepatic PAS staining showed great consumption of hepatic glycogen during fetal cardiac intervention, which could be re-accumulation in following gestation.
Conclusion: Intra-utero fetal cardiac intervention in near-term fetal lambs is of feasibility in some extend, further research needed to be done about this question.
706. Neuroprotective effects of acid-sensitive ion channel inhibitor psalmotoxin-1 in selectively vulnerable putamen after hypoxia-ischemia in newborn piglet
Z. Yang1, E. Carter1, L. Marttin2 and R. Koehler1
1ACCM; 2Pathology, Johns Hopkins University, Baltimore, Maryland, USA
Background and purpose: Acidosis is thought to play an important role in cerebral ischemic injury. Psalmotoxin-1 (PcTX), an inhibitor of Ca2+-permeable acid-sensing ion channel 1a (ASIC1a), has been shown to reduce infarct volume in adult focal ischemia models.1 Here, we tested the hypothesis that PcTX attenuates neuronal damage in a global model of neonatal hypoxic-ischemic (H-I) encephalopathy.
Methods: Anesthetized piglets were subjected to 40 mins hypoxia (9.8% O2) followed by asphyxic cardiac arrest. Groups received intracereboventricular injections of artificial cerebrospinal fluid (aCSF), PcTX (1 μg in 50 μL aCSF) 20 mins before H-I (PcTX pre-treatment) or PcTX 30 mins after H-I (PcTX post-treatment). To contrast the role of ASIC1a with NMDA receptors in this model, another group received an intravenous injection of the NMDA antagonist MK-801 (1 mg/kg) 20 mins before H-I. Animals were survived 4 days for histology or 3 h for measurements of nitrotyrosine immunoreactivity, as a marker of protein nitration, and protein carbonyl formation, as a marker of oxidative modification of proteins.
Results: At 4 days of recovery from H-I, viable neurons in putamen were markedly increased from 20±5% (±s.d.; n = 6) of sham surgical controls with aCSF injection to 47±10% (n = 7) with PcTX pre-treatment, which was similar to the 45±18% value obtained with MK-801 pre-treatment (n = 7). However, no significant protective effect was found with PcTX post-treatment (27±17%, n = 7). Pre-treatment with PcTX also resulted in improved recovery of neurological deficits on day 4 (aCSF = 40±12; PcTX = 19±13; maximum = 154). Arterial blood pressure, blood gases, and temperature were similar among groups. Immunoblots (n = 6 per group) of putamen at 3 h of recovery indicated that PcTX pretreatment reduced ischemic-induced protein nitration (PcTX 1.37±0.21 versus aCSF 2.4±0.30 fold of sham+aCSF, P<0.05) and protein carbonyl formation (PcTX 1.37±0.21 versus aCSF 1.65±0.27 fold of sham+aCSF, P<0.05).
Conclusion: PcTX pre-treatment decreased nitrative and oxidative damage and protected striatal neurons from ischemic death in immature brain. The role of ASIC1a channels appears to be at least as important as that of NMDA receptors during H-I. In contrast to efficacy after reperfusion in adult transient focal ischemia,1 post-treatment with PcTX did not show obvious protection, possibly because of early recovery of tissue acidosis after resuscitation in neonatal hypoxia-ischemia.
812. Regional pattern and time course of injury in near term rabbit fetal hypoxia-ischemia
A. Drobyshevsky, M. Derrick, X. Ji, L. Yu and S. Tan
Pediatrics, NorthShore University Healthcare System Research Institute, Evanston, Illinois, USA
Introduction: Premature infants with periventricular leucomalacia are at high risk to develop cerebral palsy. However, the type and site of lesions varies with gestational age. In late gestation, a basal ganglia type of injury often occurs with extrapyramidal clinical manifestations.
Objective: To reveal patterns of regional injury in near-term fetal brain and correlate it with the development of ensuing motor deficits in a rabbit model of fetal hypoxia-ischemia (Derrick et al, Stroke, 2007;38(2 Suppl):731–5).
Methods: New Zealand White rabbits at 29 days gestation (E29, 90% gestation) were subjected to 32 mins uterine ischemia that results in global fetal-ischemia (H-I). Fetuses were imaged serially in utero on a 3T magnet, delivered by C-section and imaged on 4.7 Bruker magnet at 4, 24, 72 h after H-I. A diffusion-weighted sequence with b = 0.800 secs/mm2 was employed and apparent diffusion coefficient (ADC) calculated. At P1 the full diffusion tensor sequence (DTI) was acquired to estimate fractional anisotropy (FA). Muscle resistance to passive stretch in the rabbit kits after near term (E29) fetal H-I was measured using a recently developed torque-displacement apparatus.
Results: Out of 31 fetuses, 12 were found dead on c-section 4 h after the H-I insult. Out of 19 survivors, 10 failed to maintain respiration 5 to 15 mins after delivery. Out of the remaining 9 surviving kits, 4 developed hypertonia of hind limbs at 72 h after H-I.
These kits had marked and persistent reduction of ADC in basal ganglia, thalamus and brainstem compared to non-hypertonic kits (Figure 1). Massive cell death occurred by 72 h after H-I in the areas of initially decreased ADC as indicated by formation of cystic lesions, tissue loss or hyperintensities on T2 and DWI-weighted images. Kits that developed hypertonia by 72 h after H-I were initially hypotonic at 24 h after H-I (Figure 2). At P1 (72 h after H-I) FA values of hypertonic kits in corpus callosum and internal capsule showed a trend to be lower than in non-hypertonic kits.
Conclusions: The biphasic response is strikingly similar to the development of human hypertonia in infants, but in a much compressed time frame. Basal ganglia, thalamus and brain stem sustain larger injury than cortex and white matter after near term H-I in rabbits. ADC at 4 h after H-I is informative to predict hypertonia in surviving kits. Development of hypertonia after near tem H-I in rabbits occurs within 72 h making the model a convenient platform to study pathophysiology of hypertonia in cerebral palsy.
816. Cerebro-vascular responses to brief global asphyxia in late gestation fetal sheep: evidence for vascular re-modelling?
A. Baburamani, D. Walker and M. Castillo-Melendez
Physiology, Monash University, Melbourne, VIC, Australia
Background: The developing brain is vulnerable to hypoxia, resulting in damage due to cerebral edema, hemorrhage and vascular leakage. In the adult brain, angiogenic responses to hypoxia/ischemia include up-regulation of vascular endothelial growth factor (VEGF) and the formation of new, immature blood vessels that are leaky and prone to rupture.1 During cerebral ischemia, the functional and structural integrity of the blood brain barrier (BBB) is rapidly destroyed;2 this may extend beyond the endothelial component of the BBB to the basal lamina, which is also vulnerable to hypoxia-ischemia.
Aim: To investigate the early cerebrovascular responses (24 to 48 h) that occur in the late gestation fetal brain following brief global asphyxia produced by umbilical cord occlusion (UCO).
Methods: At 124 to 127 days gestation (term = 147 days), singleton fetal sheep underwent surgery for catheterization and placing an inflatable cuff around the umbilical cord. At 132 days gestation, UCO was produced for 10 mins. Fetal brains were collected at 24 (n = 5) and 48 h (n = 5) after UCO following transcardial perfusion with 4% paraformaldehyde. Control fetuses (n = 5) underwent a sham UCO. Immunohistochemistry was carried out on 10 mm sections using mouse monoclonal anti-VEGF, rabbit monoclonal anti-Ki67 (proliferative marker) and polyclonal rabbit anti-laminin (an important component of the basal lamina) primary antibody and visualized with metal-enhanced diaminobenzydine. VEGF and Ki67 immunoreactivity (IR) was visualized using Image J. Laminin was quantified with AdobePhotoshop7.
Results: Following UCO, a 10 to 20 fold increase in VEGF-IR was seen at 24 and 48 h in the subventricular zone (SVZ), striatum terminalus (ST), caudo-putamen, cortex, white matter and corpus callosum (CC). Endothelial cell proliferation (Ki-67) showed a 8 to 18 fold increase at 24 h in the ST, cortex and caudo-putamen and was maintained at 48 h. The SVZ showed a 300 fold increase at 24 h which was still elevated at 48 h. The amount of laminin present in blood vessels, quantified by densitometric analysis revealed no change in most brain regions at either time point. However, in the CC, SVZ and ST a marked decrease in the number of blood vessels/mm2 was seen at 24 and maintained at 48 h after UCO.
Conclusions: Brief, global asphyxia results in the up-regulation of VEGF and endothelial cell proliferation, adaptive responses known to result in the formation of new blood vessels. A wide-spread and sustained increase was observed at 24 and 48 h following UCO, suggesting that the VEGF system is induced by brief asphyxia in the late gestation fetal sheep, leading to endothelial cell proliferation. The density of laminin-IR blood vessels/mm2 was either unchanged or reduced in the brain regions examined suggesting that growth of new vessels had not taken place by 24 or 48 h following UCO. However, as degradation of cerebral microvascular laminin reportedly occurs following brain hypoxia, further investigation into alternative vascular markers as well as assessment of the structural stability of the basal lamina (using collagen type IV and fibronectin) following fetal hypoxia needs to be carried out.
853. A broadband multi-channel NIRS system for measuring focal ischemic brain injury in newborns
M. Diop1,2, J. Elliott1,2, K. Tichauer1,2, T.-Y. Lee1,2,3 and K. St. Lawrence1,2
1Imaging Division, The Lawson Health Research Institute, St. Joseph's Health Care; 2Department of Medical Biophysics, University of Western Ontario; 3Imaging Research Laboratories, Robarts Research Institute, London, ON, Canada
Objectives: Near-infrared spectroscopy (NIRS) is a promising technique for assessing brain function in newborns. We have previously developed a method for measuring cerebral blood flow (CBF) using broadband NIRS.1 However, broadband systems typically have one detector, which limits their clinical applicability. To address this limitation, we developed a multiplexing approach based on electronically controlled mechanical shutters to expand the detection capabilities to eight channels. This work was intended to demonstrate that broadband NIRS can detect focal ischemic brain injury caused by injecting Endothelin-1 (ET-1). For validation, CBF was measured by computed tomography (CT) perfusion.2
Methods: Four newborn piglets were studied: 2 females, 2 males; median age 1.5 days. Animals were anaesthetized using isoflurane and physiological parameters were maintained within normal limits throughout the experiments. Piglets were placed in a CT scanner with the NIRS emission optode and four detection optodes positioned on the head. All detectors were placed 3 cm from the emitter, spanning an arc of 90°. Two sets of CBF measurements were obtained by CT2 and NIRS1 before and following the infusion of ET-1 (5 μmol in 50 mL of water) into the cerebral cortex.3
All data are presented as means±s.e.m. For both NIRS and CT, the two CBF measurements acquired prior to and following ET-1 infusion were averaged to generate single pre- and post-injection values. Comparison between the techniques was conducted for two regions: the site of ET-1 infusion (injury region) and a region remote from infusion (healthy region). Significant differences were determined by paired Student's t-test (P<0.05) corrected for multiple comparisons.
Results: Baseline measurements showed no statistical differences between CBF measured by the four NIRS optodes and average CBF from the corresponding regions in the CT perfusion images. Infusion of ET-1 caused a general reduction in global CBF with the magnitude of the reduction diminishing with distance from the infusion site. With NIRS, a significant decrease was found only in the injured region (51±9%) and not in the healthy region (14±9%). Significant decreases in CBF for the corresponding regions (55±5 and 23±8%, respectively) were measured by CT.
Conclusion: Regional CBF measured by NIRS and CT were in good agreement with each other and both techniques measured approximately a 50% reduction in CBF in the injury region following ET-1 infusion. This study shows that the multi-channel broadband NIRS system can reliably measure regional CBF. It would be possible to also measure the cerebral metabolic rate of oxygen by combining measurements of CBF and cerebral deoxy-hemoglobin,4 making this mobile NIRS apparatus a valuable tool for bedside monitoring of critically-ill neonates.
1003. A new swim test quantifies movement deficits in perinatal rabbits after fetal hypoxia-ischemia
M. Derrick1,2, A. Drobyshevsky1, X. Ji1, Y. Yang1, R. Silverman3, H. Ji3 and S. Tan1,2
1Pediatrics, Northshore University Healthsystems; 2Pediatrics; 3Chemistry, Northwestern University, Evanston, Illinois, USA
The relationship of movement of different muscle groups has not been quantified before in the newborn period. Cerebral palsy (CP) which often occurs as a result of perinatal hypoxia-ischemia (H-I) is categorized to different types depending on clinical presentation, involvement of brain region and extent of involvement. In order to test involvement of different brain regions, this study investigates individual and multi-joint involvement in a rabbit model of CP. Pregnant rabbits at 70% gestation were subjected to 40-mins uterine ischemia. Newborn rabbit kits were subjected to a swim test at 5 time points over the first 1 to 11 days of life. H-I kits were divided into hypertonic and non-hypertonic groups based on muscle tone at birth. The ranges and velocity of angular movement of the upper and hind-limb joints (wrist, elbow, shoulder, ankle, knee and hip) during supported swimming were determined. Severely impaired (hypertonic) animals have significantly reduced range and angular velocity of joint motion which do not improve over time. The non-hypertonic group showed deficits in wrist and hind-limb movements that were not evident on prolonged observation. Preventive treatment with an inhibitor of neuronal nitric oxide synthase decreased the incidence of severely impaired kits; the non-hypertonic kits showed a different pattern of swimming. Supported swimming allows easy quantification of limb and joint motion in the principal plane of movement in the absence of weight bearing and decreases the need for control of balance. Identification and quantification of milder deficits allows mechanistic studies in the causation of H-I injury as well as estimation of recovery from therapeutic agents.
1009. Neuronal nitric oxide synthase level is a critical determinant of neuronal death following fetal hypoxia-ischemia
L. Yu1, A. Drobyshevsky1, X. Ji1, M. Derrick2 and S. Tan2
1NorthShore University Healthcare System Research Institute; 2Northshore University Healthsystems, Evanston, Illinois, USA
Background: We have previously published that apparent diffusion coefficient (ADC) by MRI in fetal hypoxia-ischemia (HI) differentiates fetuses that are destined to be hypertonic postnatally from those that do not have motor deficits.1 We have also shown that neuronal nitric oxide synthase (nNOS) inhibitors prevent the motor deficits and perinatal deaths if given prior to HI.2
Objectives: We hypothesized that nNOS is a critical determinant of neuronal death that results in hypertonia. We tested a novel methodology combining ADC prediction with high speed flow cytometric sorting of cells and quantitative RTPCR.
Methods: Rabbit dams at E25 were subjected to 40 mins uterine ischemia in the 3T magnet and serial ADC measurements obtained. ADC threshold was used to categorize fetuses destined to be hypertonic. After 30 mins of reperfusion, fetal brains were removed and dissociated into single cell suspensions. Cells were stained by Rhodamine 123 and propidium iodide (PI) and sorted. Sorted cells were then assessed for nNOS gene expression by quantitative RTPCR.
Results: Cells were categorized into dead, injured and healthy cells based on Rhodamine and PI staining. Dead cells (PI+ and rhodamine−) showed higher nNOS expression than injured (PI+ and rhodamine+) or healthy cells (PI− and rhodamine+). In fetuses destined to be hypertonic, nNOS expression was even higher.
Conclusions: We conclude that nNOS is a critical determinant of cell death. We speculate that a certain threshold level of nNOS tips the fetus from a low risk to a high risk fetus for subsequent hypertonia.
1016. Prostaglandin E2-EP4 receptor agonist protects neonatal brain from hypoxia-ischemia
H. Taniguchi, C. Anacker, X. Liang, Q. Wang, F. Sabar and K. Andreasson
Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, California, USA
Objectives: The pathogenesis of neonatal hypoxic-ischemic encephalopathy (HIE) involves increased production of prostaglandins.7 It is known that two Gαs-coupled receptors, prostaglandin D2-DP1 receptor and prostaglandin E2-EP2 receptor, mediate neuroprotection in vivo in models of brain ischemia1–3,5,7). In this study, we sought to clarify the role of the prostaglandin E2-EP4 receptor, another Gαs-coupled receptor, in a model of neonatal brain ischemia.
Methods: The neonatal HIE model was performed as previously described.4,6 Briefly, the left common carotid artery of postnatal day 7 Sprague-Dawley rat pups was ligated and pups were placed in hypoxic chamber for 100 mins. Either ONO-AE1–329 (selective EP4 agonist; Ono Pharmaceuticals, Osaka, Japan) or vehicle were injected subcutaneously before and/or after hypoxia. At 24 h after the end of hypoxia, the pups were sacrificed and brains were sliced and stained with 2,3,5-Triphenyltetrazolium chloride. Infarct area and infarct volume were measured using Image J (NIH, Bethesda, MD).
Results: Immunostaining of the EP4 receptor shows that the EP4 receptor is expressed in neurons and endothelium of cerebral cortex, striatum, and hippocampus. Stimulation of EP4 receptor with ONO-AE1–329 protected organotypic slice CA1 hippocampal neurons from N-methyl-d-aspartate (NMDA) toxicity or oxygen-glucose deprivation.
The infarct size was in HIE model significantly reduced in the EP4 agonist treated group compared with the vehicle treatment group. The EP4 agonist was most effective when administered before hypoxia.
Conclusions: We found that the EP4 agonist, ONO-AE1–329 reduced the infarct size in our neonatal HIE model. Taken together, these findings indicate that in vitro and in vivo, stimulation of the EP4 receptor leads to neuronal protection. Early administration of this EP4 agonist may provide a novel approach to protect the neonatal brain from hypoxic-ischemic insult.
1017. Cerebral palsy phenotype in rabbits modeling progression from partial to complete placental insufficiency
X. Ji1, M. Derrick2, L. Yu1, A. Drobyshevsky1, A. Liu1 and S. Tan3
1NorthShore University Healthcare System Research Institute; 2Pediatrics; 3Peditrics, NorthShore University Healthcare System Research Institute, Evanston, Illinois, USA
Objective: Fetal hypoxia-ischemia is one of the important causes of cerebral palsy in children. The clinical scenario of abruptio placenta is a significant cause of fetal hypoxia-ischemia. Abruptio placenta progresses from partial to total abruptio in a variable period of time. To model the progression of partial to complete placental insufficiency, we modified our previously published animal model mimicking cerebral palsy.1 The objective was to make the animal model a more clinically relevant model.
Methods: Pregnant New Zealand White rabbits at 70% gestation (E22) were subjected to partial uterine ischemia for 30 mins by inflation of a balloon catheter in the distal aorta using 75 μL. Fetal bradycardia was noted within 5 mins of the balloon inflation. Then complete uterine ischemia was implemented for 40 mins by further inflation of balloon to 250 μL. The balloon was then deflated, the aortic catheter removed, wound sites sutured, and the pregnant dams were allowed to survive. The dams spontaneously gave birth at term gestation (E31.5 days). The newborn kits were subjected to a battery of neurobehavior tests on the first postnatal day.
Results: Of the 52 kits born to 8 dams, there were 25 stillbirths or fetal deaths (48%). 85% of the survivors manifested hypertonia based on a modified Ashworth scale. Based on the total neurobehavioral score of the survivors, there were 15% normal, 33% mild and 52% severely affected kits. Weights of survivors were 56±2 and dead kits were 35±3 G.
Conclusions: This outcome is slightly more severe than our previous model. We conclude that this modified model more closely mimics the clinical situation of the human mother who undergoes first partial then complete abruptio placenta. The rabbit kits show a spectrum of injury suggesting that some rabbits are intrinsically more vulnerable than others. The incidence of hypertonic survivors may be an underestimate because most of the stillbirths/fetal deaths may be hypertonic if we were able to resuscitate them. Furthermore, this model is suitable for testing neuroprotectants administered in the antenatal period.
1032. Articulation characteristics of developmental verbal apraxia
V. Djordjevic1 and S. Golubovic2
1Institute for Experimental Phonetic and Speech Pathology, Belgrade; 2Speech and Language Pathology, Faculty of Special Education and Rehabilitation, University of Belgrade, Belgrade, Serbia
Introduction: Verbal apraxia or apraxia of speech is impaired ability to execute voluntarily the appropriate movements for articulation of speech in absence of parallysis, weakness, or incoordination of the speech musculature. Errors are inconsistent, involuntary speech is better than voluntary speech, errors include substitution, repetition, simplification, distortion, and addition. The errors increase with complexity and word length.
Method: We researched articulation ability in 15 children with verbal apraxia, 15 with developmental dysphasia and 15 with developmental phonological disorders (age 4 to 5) with Test of Articulatin sounds in Serbian Language.
Results: Analys of results showed us that substitution of sounds have 25.5% of children with dysphasia, 12.8% with verbal apraxia and 10.4% with phonological disorders. Distortion of sounds have 20.4% of children with verbal apraxia, 13.7% with phonological disorders and 6% with dysphasia. Omision of sounds have 7.7% of children with verbal apraxia 7.3%, with dysphasia and 3.3% with phonological disorders. Children with developmental dysphasia and verbal apraxia must recent have substitution and distortion of affricates, fricatives, laterals, nasals, plosives, and vowels, while children with phonological disorders have more recent substitution of nasals than plosives. The largest frequency articulation of disorders consonant (substitution and distortion) is in children with dysphasia and than with verbal apraxia and phonological disorders. Between children with phonological disorders and dysphasia and verbal apraxia there is no difference in number of omision consonant clusters, distortions CVCV, and CVC.
Conclusion: Affricates and fricatives tend, as classes, to be more often in error than plosives, laterals, nasals, and vowels, although order varies with the position in utterance. Consonant errors are more likely than vowel errors, some children may make no more consonant errors than vowel errors.
1038. Ability of speech production and comprehension in children with developmental language disorder
L. Zecevic-Filipovic1 and S. Golubovic2
1Kindergarten Tivat, Tivat, Montenegro; 2Faculty of Special Education and Rehabilitation, University of Belgrade Education and Rehabilitation, University of Belgrade, Belgrade, Serbia
Objectives: Developmental language disorder (DLD) is defined as a failure of normal language development in a child with normal nonverbal intelligence, no other neurologic or psychiatric disease, and no hearing loss. Children with this form of DLD exibit predominant receptive and expressive impairments of phonology and grammar. Dysphasic children are also characterized by speech production errors including misproduction and omission. As was found for speech perception, dysphasic children are preferentially impaired for producing speech sounds characterized by rapid changes.
Method: Sample of examinees includes 30 children, age of 5, 6 years, 15 girls and 15 boys. All of there are tested with Reynell scale of language development, which is standardizes by Croatia and Serbian language of examinees. Test examinees level of ability of understanding speech reception and speeech production. Scale of speech understanding is composite part of test which examine structure, vocabulary and content, and on the basis of summarized points and equivalent age it is calculated deviation related to arithmetic mean. It is expressed in standard deviation and with that method is obtained standard evaluation which is equivalent to language development for certain calendar age.
Result: Result of examined sample of children shows small difference between boys and girls (of same calendar age = 5, 6) on tasks of Scale of speech understanding. But, it is noticed significantly better understanding in 6 boys and 4 girls, than acquirement on scale of speaking expression, and especially on part related to structure and content of speech, and which examine semantic and syntatic level.
Conclusion: Developmental expressive language disorder with impaired ability to express oneself and difficulty with vocabulary, complex sentences and recall of words. In severe cases DLD may persist until school-age, and the majority of affected children will develop additional learning disabilities or reading deficits.
References
1.
Bonnin. Ultrasound in Med & Biol, 2008;34:913–22.
2.
ZengJYangGYYingWKellyMHiraiKJamesTLSwansonRALittL. Pyruvate improves recovery after PARP-1-associated energy failure induced by oxidative stress in neonatal rat cerebrocortical slices. J Cereb Blood Flow Metab2007;27:304–15.
LentiLDomokiFKisDHegyiOTothGKBusijaDWBariF. Pituitary adenylate cyclaseactivating polypeptide induces pial arteriolar vasodilation through cyclooxygenase-dependent and independent mechanisms in newborn pigs. Brain Res2007;1165:81–8.
7.
MäkikallioKMcElhinneyDBLevineJC. Fetal aortic valve stenosis and the evolution of hypoplastic left heart syndrome: Patient selection for fetal intervention. Circulation2006;113:1401–52.
8.
GalindoAGutiérrez-LarrayaFVelascoJM. Pulmonary balloon valvuloplasty in a fetus with critical pulmonary stenosis/atresia with intact ventricular septum and heart failure. Fetal Diagn Ther2006;21:100–43.
9.
TworetzkyWMarshallAC. Fetal interventions for cardiac defects. Pediatr Clin North Am2004;51:1503–13, vii4.
10.
Wilkins-HaugLEBensonCBTworetzkyW. In-utero intervention for hypoplastic left heart syndrome—a perinatologist's perspective. Ultrasound Obstet Gynecol2005;26:481–65.
11.
IkaiARiemerRKRamamoorthyC. Preliminary results of fetal cardiac bypass in nonhuman primates. J Thorac Cardiovasc Surg2005;129:175–816.
12.
FokWYLeungTYTsuiMH. Fetal hemodynamic changes after amniotomy. Acta Obstet Gynecol Scand2005;84:166–97.
13.
WoodCEGirouxD. Central nervous system prostaglandin endoperoxide synthase-1 and -2 responses to oestradiol and cerebral hypoperfusion in late-gestation fetal sheep. J Physiol2003;549(Part 2):573–818.
14.
EghtesadyPSedgwickJASchenbeckJL. Maternal-fetal interactions in fetal cardiac surgery. Ann Thorac Surg2006;81:249–559.
15.
MeuliMMeuli-SimmenCHutchinsGM. In utero surgery rescues neurological function at birth in sheep with spina bifida. Nat Med1995;1:342–710.
16.
McClaineRJUemuraKMcClaineDJ. A description of the preterm fetal sheep systemic and central responses to maternal general anesthesia. Anesth Analg2007;104:397–406.
17.
Pignataro. Brain2007;130:151–8.
18.
KaurC. Glia2006;54:826–39.
19.
del ZoppoGJHallenbackJM. Thromb Res2000;98:73–81.
20.
Brown. Pediatr Res2002;51:564.
21.
Cenic. AJNR Am J Neuroradiol1999;20:63.
22.
Huges. J Neuropathol Exp Neurol2003;62:1276.
23.
Tichauer. J Cereb Blood Flow Metab2006; 26:722.
24.
DrobyshevskyA. Annals of Neurology,2007;61:307–14.
25.
JiH. Annals Neurol (in press).
26.
LiJLiangXWangQBreyerRMMcCulloughLAndreassonK. Misoprostol, an anti-ulcer agent and PGE2 receptor agonist, protects against cerebral ischemia. Neurosci Lett2008;438:210–5.
27.
LiuDWuLBreyerRMattsonMPAndreassonK. Neuroprotection by the PGE2 EP2 receptor in permanent focal cerebral ischemia. Ann Neurol2005;57:758–61.
28.
McCulloughLWuLHaugheyNLiangXHandTWangQBreyerRMAndreassonK. Neuroprotective function of the PGE2 EP2 receptor in cerebral ischemia. J Neurosci2004;24:257–68.
29.
RiceJEIIVannucciRCBrierleyJB. The influence of immaturity on hypoxic-ischemic brain damage in the rat. Ann Neurol1981;9:131–41.
