Selected Abstracts 30.3/B22 - The targeting of acetylcholinesterase and butyrylcholinesterase to the cell surface in brain depends upon the association of the catalytic subunit with prima in the endoplasmic reticulum *E. KREJCI1, V. BERNARD1, A. DOBBERTIN1, K. DEMBELE2; 1UFR Biomedicale, INSERM U686, Paris, FRANCE, 2Médecine interne, Département de physiologie, université du Manitoba, MB, CANADA. It is well established that acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) are organized into multiple molecular forms ranging from monomers to tetramers. Part of them is linked through the collagen Q (ColQ) to the basal lamina or through the small transmembrane protein (PRiMA, Proline Rich Membrane Anchor, PRiMA) to the cell membrane. Both enzymes contain two autonomous domains, a divergent catalytic domain and a common tetramerization domain (WAT Tryptophan Amphiphilic Tetramerization). ColQ and PRiMA contain a proline rich sequence (PRAD) that organizes AChE and/or BChE into tetramers through the remarkable interaction of one PRAD and four WATs. In cell culture, all oligomers are secreted into the medium or linked to the plasma membrane. Several experiments suggest that AChE is organized into tetramers in the endoplasmic reticulum, part of which binds ColQ or PRiMA in the Golgi apparatus. Recent experiments have suggested that the PRAD/WAT interaction occurs in the endoplasmic reticulum. Where is AChE/PRiMA complex assembled? To address this question in vivo, we analyzed the localization of AChE in the striatum of mutant mice in which PRAD domain of PRiMA or WAT domain of AChE 1 are removed by homologous recombination. As expected, AChE is not accumulated at the cell surface and the AChE labelling is restricted to the cell body in PRiMA or WAT mutants. Moreover, we quantified the localization of AChE protein at EM level. 1) The quantity of AChE is higher in the endoplasmic reticulum of PRiMA mutant than in wild type or WAT mutant. 2) AChE labelling is completely absent in the Golgi apparatus of PRiMA mutant. These observations strongly support the statement 1) the oligomerization of AChE occurs in the ER through the specific interaction of PRAD with WAT; 2) the WAT domain is a strong retention signal of AChE in the endoplasmic reticulum in vivo. 1 Camp S, Zhang L, Marquez M, de la Torre B, Long JM, Bucht G, & Taylor P. Acetylcholinesterase (AChE) gene modification in transgenic animals: functional consequences of selected exon and regulatory region deletion. Chem Biol Interact. 157-158:79-86. 2005 55.10/Q16 - Characterization of voltage-gated sodium currents in postnatal mouse spinal motoneurons after short-term culture *K. P. CARLIN, J. LIU, L. M. JORDAN; Dept. Physiol, Univ Manitoba, Winnipeg, MB, CANADA. Voltage-gated sodium channels are fundamental to the functioning of excitable cells. As these channels are influenced by various cellular components, it is preferable to study their functioning in their native cellular background as opposed to heterologous expression systems. In spinal motoneurons these channels are mainly localized to areas of the plasma membrane at some electrotonic distance from the soma - such as the initial segment and Nodes of Ranvier. This fact makes difficult the characterization of the biophysical properties of these channels and kinetics of the currents. Even though the culturing of embryonic motoneurons has yielded much information on these currents, developmental changes in the main sodium channel subtypes expressed in these cells limits the ability to extrapolate these results to mature systems. To circumvent these problems we isolated postnatal motoneuronal somata using culture techniques. Motoneurons (P1 - P12 mice) were pre-labeled with Fluoro-gold then dissociated and subjected to short-term (18-24 hrs) culture conditions designed to promote survival but not process formation. Sodium currents were then recorded from labeled sphere-like cells (n = 45). These currents activated in a voltage dependent manner with a half maximal activation voltage (V1/2) of -22 +/- 4.7mV (mean +/- SD) and slope of 4.4 +/- 0.7. The V1/2 of inactivation averaged -53 +/- 6.5mV with a slope value of 6.4 +/- 0.9. Using a 20ms paired-pulse protocol the current was seen to recover from inactivation with fast (tau = 17 +/- 8ms) and slow (tau = 140 +/- 125ms) phases. The time required for the channels to inactivate in response to a 20mV depolarization was fit well with a single exponential and at 2 seconds 48 +/- 20% of the current still remained. The ability of this current to follow various stimulation frequencies was also tested. Modest frequency-dependent inhibition of the current was observed at the lowest frequency tested (1Hz ; 0.93 +/- 0.04; pulse 20/ pulse 1) but increased to where only 31 +/- 1.1% of the current remained at 30 Hz (Vh = -70mV x 20ms). Some age-dependent trends were noted in the data with more mature cells tending to recover from inactivation faster and to better follow high frequency stimulation. These experiments provide insight into the postnatal changes in motoneuronal firing properties and provide a baseline of sodium channel functioning from which the mechanisms of neurotransmitter modulation can be studied. 55.11/R1 - Hyperpolarization of voltage threshold enhances motoneuron excitability during fictive scratch in the cat *K. E. POWER, D. S. WANG, D. A. MCCREA, B. FEDIRCHUK; Physiology, University of Manitoba, Winnipeg, MB, CANADA. Motoneuron excitability is enhanced by a lowering of the voltage threshold (Vth) for action potential initiation during brainstem evoked fictive locomotion in the decerebrate cat (Krawitz et al., 2001). Vth hyperpolarization is also seen in neonatal rat spinal neurons during electrical stimulation of the brainstem (Gilmore and Fedirchuk, 2004) and during bath application of the mononamines serotonin or noradrenaline (Fedirchuk and Dai, 2004). Whether Vth hyperpolarization occurs during other motor behaviours or is dependent upon brainstem stimulation in the cat is not known. Experiments were conducted on decerebrate adult cats in which fictive scratch was induced in the ipsilateral hindlimb by applying a small piece of curare soaked cotton on the left C1 dorsal root entry region, followed by light stroking of the left side of the face. Measurements of the Vth of the first spike evoked by intracellular injection of depolarizing ramp currents or square wave pulses were compared during control and fictive scratch in discontinuous current clamp mode. Recordings were made from 16 antidromically identified extensor motoneurons in intact cats. Vth hyperpolarized in 15 of 16 extensor motoneurons (mean hyperpolarization -6.2 ± 3.2 mV; range -1.6 to -11.9 mV) with the onset of fictive scratch. In one cat a complete transection of the spinal cord at C1 was made. In 5 of 8 extensor motoneurons in this preparation fictive scratch resulted in Vth hyperpolarization. The results show that fictive scratch induces a hyperpolarization of the Vth for action potential initiation in both intact and spinal cats. Vth hyperpolarization is not dependent upon electrical brainstem stimulation and may accompany a variety of rhythmic motor behaviours. 55.14/R4 - Spike frequency adaptation of rat hindlimb alpha motoneurons *D. C. BUTTON, J. M. KALMAR, K. GARDINER, F. CAHILL, P. F. GARDINER; Physiology, Univ Manitoba, Winnipeg, MB, CANADA. Spike frequency adaptation (SFA) refers to a decrease in motoneurone (MN) firing frequency (FF) during sustained excitation which exceeds the threshold for rhythmic firing (ThrR). SFA is comprised of three phases (initial, early, and late) and is likely the result of several physiological mechanisms. This complexity has made it difficult to quantify SFA. The purpose of this study was to resolve two methodological issues surrounding the quantification of MN SFA: 1) to determine if SFA is current-dependent and, 2) to develop an estimate of SFA that is sensitive to the wide range in adaptation patterns between MNs. Sciatic motoneurones were impaled by glass microelectrodes in the L3-L4 region of the spinal cord of anaesthetized (ketamine/xylazine) Sprague-Dawley rats to record electrophysiological properties. SFA was measured in response to 3, 30-s square wave current injections that were 1.5, 3.0, and 5.0 nA > ThrR. In this study, the decline in the area under the frequency-time curve was used to estimate SFA using the following procedure. For each cell, the number of spikes per second was normalized such that the minimum FF over the 30-s recording was always 5 Hz. An ANOVA revealed a significant current x time interaction (F=4.16, p<0.001) such that the normalized FF (FFnorm) was higher in the 5.0 nA > ThrR trial compared to 1.5 nA and 3.0 nA > ThrR trials (p<0.001), but only in the first 2-s of current injection. Thereafter, there was no difference in FFnorm between the three current trials. The area in each 2-s bin of the FFnorm-time curve was then expressed relative to total area. We also measured other MN active and passive properties and correlated them to SFA. SFA expressed as a ratio of the 13th bin area (24-26s) to the 2nd bin area (2-4s) significantly correlated with estimated persistent inward current amplitude (r = 0.69, p < 0.005) and rheobase current (r = -0.52, p < 0.05). In conclusion, after the first 2-s of current injection, any current up to 5 nA > ThrR will derive the same SFA, eliminating the time-consuming process of injecting multiple currents. Furthermore, SFA quantified using frequency-time area correlates well with other MN properties. Thus, this estimate of SFA may be used in future studies to characterize changes in SFA in response to altered activity and other perturbations. 55.18/R8 - Ion channel subunit and trkB gene expression changes in rat motoneurons as a result of treadmill exercise *H. JI, K. GARDINER, P. F. GARDINER; Physiology, Univ Manitoba, Winnipeg, MB, CANADA. Our previous studies have shown that there are significant changes in the electrophysiological properties of rat lumbar a-motoneurones (MNs) following increased voluntary wheel exercise and treadmill training (12-16weeks). However, little is known as to the mechanisms of adaptation of MN properties in response to alterations in neuromuscular usage. The purpose of this study was to determine if exercise-induced changes occur in expression of genes coding for fast sodium and delayed-rectifier potassium channel subunits, and the neurotrophin receptor trkB. Twenty-four adult female Sprague-Dawley rats (250-300g) were randomly separated into four groups: ventral lumbar spinal cord exercise (VE), ventral lumbar spinal cord control (VC), laser capture microdissection exercise (LCME), and laser capture microdissection control (LCMC). The exercise groups were treadmill-exercised at a speed of 15 m/min while the grade increased from 0 to 10% and the duration increased from 15 to 30 min, over the 5 days of exercise. All rats were sacrificed within 1 h of the last exercise bout. In groups VE and VC, total RNA was extracted from the ventral part of lumber spinal cord. In groups LCME and LCMC, Toluidin Blue staining was used to visualize MNs in 15-µm sections, from which 50 MNs were harvested for each cord using LCM. Real time RT-PCR was carried out using GAPDH and 18S as reference genes. Although the expression of NaV1.6, KV1.1, CaV1.3, SK2, and trkB in group VE was greater by 59.65%, 32.7%, 2.33%, 24.6%, and 30.4% compared to group VC, these differences were not significant. The gene expression of NaV1.6, KV1.1, and trkB in MNs in group LCME was significantly greater than in group LCMC (3.2-, 2.8-, and 4.6-fold, p = 0.032, 0.016, and 0.001, respectively). In conclusion, MNs respond to 5 days of daily exercise by increases in the expression of genes coding for sodium and potassium channel subunits and neurotrophin receptors. Changes are not evident when analysing ventral spinal cord, but are evident when analysing “enriched” samples of MNs harvested by laser microdissection. Treadmill training thus regulates the gene expression of ion channel subunits and neurotrophin receptors in MNs, and this may play a role in changes in MN properties seen previously. 58.7/T20 - Stress differentially effects TLR-mediated cytokine responses in vivo *B. J. MACNEIL1, D. M. NANCE2; 1Physical Therapy, Univ Manitoba, Winnipeg, MB, CANADA, 2Susan Samueli Center for Integrative Medicine, UCI, Irvine, CA. Although stress is known to alter immune function, the nature of the effect is difficult to predict since both reduced and enhanced immune responses are reported. Direct comparison across studies is limited in part due to use of different immune stimulants. The availability of specific toll-like receptor (TLR) ligands makes it possible to standardize the immune stimulant. This study tested the hypothesis that stress alters innate immune responses in a TLR-dependent manner. Male Sprague-Dawley rats were exposed to acute stress (15 minutes intermittent footshock) followed by iv injection of ligands specific for TLR2 (Pam3CSK4), TLR3 (poly I:C) or TLR4 (ultrapurified LPS). Animals were sacrificed 1 or 3 hours after injection to determine splenic cytokine expression relative to unstressed rats. Northern blotting was used to determine cytokine mRNA expression directly from spleen homogenates. One hour after TLR2, TLR3 or TLR4 stimulation mRNA for TNF-alpha, IL-1beta, MCP-1, IL-18, MIP-1 and RANTES was detected in the spleen. Despite a similar pattern of cytokine expression across all TLR ligands, the effects of stress were selective for TLR2 and TLR4. Specifically, stress dramatically inhibited TLR2- and TLR4-induced TNF mRNA expression (p<0.05 for both) whereas stress had no effect on TLR3 stimulated TNF production. A similar pattern of cytokine expression was seen three hours after TLR stimulation with the additional detection of IFN-gamma mRNA. Again, stress reduced levels of TNF mRNA following TLR4 stimulation (P<0.05) but did not effect TNF expression after TLR3 stimulation. Despite the distinct effects of stress on TNF mRNA responses, IFN-gamma mRNA expression was similarly inhibited following either TLR4 or TLR3 stimulation (p<0.05 for both). In contrast, IL-18 mRNA levels tended to be enhanced following TLR3 stimulation at the 1 hour (p=0.064) and 3 hour (p=0.08) intervals whereas TLR4-mediated IL-18 responses were unaffected by stress. These data show that stress can act in a TLR-dependent manner to produce cytokine-specific alterations in innate immune responses. Understanding the mechanisms by which stress selectively alters cytokine subsets may provide avenues for selective manipulation of cytokine responses. 83.23/LL83 - BNIP3 and EndoG define a novel neuronal cell death pathway in hypoxia and stroke *J. KONG, Z. ZHANG, X. YANG, S. ZHANG, X. MA; Human Anatomy and Cell Science, University of Manitoba, Winnepeg, MB, CANADA. Caspase-independent neuronal apoptosis is widely observed in hypoxia and stroke. However, little is known about the molecular pathways for this atypical form of apoptosis. Previously, it has been reported that the gene BNIP3 when expressed induces cell death in a necrotic-like manner in non-neural cell lines. This gene-regulated cell death pathway involves opening of the mitochondrial permeability transition pore without caspase activation and cytochrome c release. Here we test the hypothesis that BNIP3 activates an atypical form of neuronal apoptosis in stroke and hypoxia, and EndoG is a mediator of the BNIP3-activated neuronal cell death pathway. Our results show: 1) BNIP3 is upregulated in brain neurons in the MCAO rat model of stroke and in primary neuron cultures exposed to hypoxia. The expressed BNIP3 is localized to mitochondria and increases opening probability of mitochondrial permeability transition pores. 2) Forced expression of BNIP3 in neurons induces mitochondrial dysfunction and cell death; 3) Knockdown of BNIP3 with a potent shRNA-expressing lentiviral vector (inhibition efficiency of 98.1%) protects against neurons from hypoxia-induced cell death. From the inhibition study we estimate the BNIP3-activated pathway accounts for at least 15-23% of hypoxia-induced neuron death. 4) Hypoxia induces mitochondrial release and nuclear translocation of endoG with a time course identical to that of hypoxia-induced BNIP3 expression. 5) Forced expression of BNIP3 induces EndoG translocation in both neurons and non-neural cells; 6) Inhibition of BNIP3 expression significantly delayed EndoG translocation. From the data, we propose a novel BNIP3 and EndoG-defined neuronal cell death pathway in hypoxia and stroke. In this pathway, BNIP3 is transcriptionally upregulated by the transcriptional factor HIF-1 in hypoxia and stroke, causes mitochondrial dysfunction and results in release of EndoG, which translocates to the nucleus and cleaves chromatin DNA leading to a caspase-independent neuronal cell death. This pathway has also been confirmed in a model of neuroexcitotoxicity and may be a common cell death pathway in neurodegenerative diseases. 92.15/NN66 - Modulation of synaptic plasticity in hippocampus in streptozotocin-diabetic rats *K. OIKAWA1, G. ODERO1, D. SMITH1,2, G. W. GLAZNER1,2, P. FERNYHOUGH1,2, B. C. ALBENSI1,2; 1Division of Neurodegenerative Disorders, St. Boniface General Hospital Research Centre, Winnipeg, MB, CANADA, 2Pharmacology & Therapeutics, University of Manitoba, Winnipeg, MB, CANADA. Type 1 diabetes mellitus is a condition in which the pancreas does not make insulin, leading to severe hyperglycemia and numerous secondary pathologies, including peripheral neuropathy and vascular dementia. Changes in CNS function appear as cognitive defects similar to those characteristic of Alzheimer’s Disease. In order to study possible changes in cellular correlates of memory encoding, such as synaptic plasticity (ie., experimental paradigms of long-term potentiation (LTP) in the CA1 hippocampus) and CREB expression/function, we used the streptozotocin (STZ) model of diabetes, in which STZ, injected i.p. kills insulin-producing pancreatic beta cells, thus rendering the animal type 1 diabetic. Three groups of adult rats were tested (controls, diabetic, and rats with subcutaneous implants that delivered insulin at 1 unit/24 hrs) where animals had STZ treatment for 8-12 weeks. This insulin treatment regime did not impact on hyperglycemia, but did provide increased insulin-dependent neurotrophic support. Our preliminary data shows that LTP was impaired in the STZ group in some cases and that paired-pulse (PP) facilitation and LTP were enhanced in the implanted group. Our initial LTP results are consistent with the results of other labs using the STZ model (Chabot et al. 1997; Kamal et al. 1999; Artola et al. 2005). However, enhanced LTP and PP facilitation from peripheral insulin delivery has not been shown before and may point to increases in the probability of glutamate release as a result of insulin administration, a mechanism that could be targeted for intervention. 128.22/B54 - Neuregulin-b1 enhances glutamate-mediated calcium influx in rat hippocampal neuronal cultures *J. S. SCHAPANSKY, G. W. GLAZNER; Pharmacology & Therapeutics, St. Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA. The neuregulin family of signaling proteins (ARIA, GGF, Her) has been implicated in several neuronal pathways, including regulation of expression of acetycholine receptors in skeletal muscle, oligodendrocyte formation, and neuronal migration and specification. One neuregulin receptor, the EGF tyrosine kinase receptor ErbB4, has been isolated to lipid rafts in neuronal synapses, indicating that neuregulin signaling may be important in synaptic transmission. Neuregulin has been implicated as a possible genetic factor in some forms of schizophrenia, a disease also associated with changes in synaptic transmission. Here we tested the hypothesis that neuregulin can modulate NMDA signaling in neurons. We observed that a 24 hour incubation of embryonic rat hippocampal neuron cultures with the ligand neuregulin-b (NRG-b1) significantly increased acute (5 min) NMDA and glutamate-mediated elevations in intracellular calcium, while a two hour treatment had no significant effect. In spite of this increase, morphological cell survival analysis revealed that NRG-b1 partially prevented excitotoxic cell death following chronic glutamate exposure (12 hours) in culture. The increased survival was partially explained by the observation that the NRG-b1-induced calcium increase did not persist, as there was no significant difference in either calcium or ROS levels in NRG vs. untreated controls after 4 hours of glutamate treatment. In order to elucidate the signaling pathways, NRG-treated neurons were exposed to various signal transduction inhibitors. Pre-incubation with MAP kinase inhibitor PD98059 or PI3Kinase inhibitor LY294002 prevented the acute NMDA and glutamate-mediated elevations in intracellular calcium induced by NRG-b1, suggesting the AKT and Erk pathways are involved in NRG-b1-induced calcium signaling. Thus, NRG-b1 may be an important regulator of glutamate-mediated synaptic transmission in CNS neurons 130.6/C9 - Characterization of persistent inward currents (PICs) in locomotor activity-related neurons of Cfos-EGFP mice *Y. DAI, L. M. JORDAN; Dept Physiol, Univ Manitoba, Winnipeg, MB, CANADA. The persistent inward currents (PICs) have been shown to play an important role in initiation of action potentials and regulation of the input/output properties of spinal motoneurons. Activation of PICs could amplify the synaptic inputs and increase neuronal excitability. Although PICs have been demonstrated in spinal interneurons in previous studies, the role of PICs in generation of rhythmic activity during locomotion and the details of PIC properties in spinal interneurons related to locomotion still remain unknown. Using cFos-EGFP transgenic mice (P6-P12) we are able to target the spinal interneurons activated by locomotion. In this study we characterized PICs in the EGFP+ neurons and investigated the modulation of PICs by 5-HT. Following a locomotor task, whole-cell patch clamp recordings were obtained in spinal cord slices (200-250 um, T12-L4) from EGFP+ neurons distributed in lamina VII, VIII and X. PICs were recorded by a family of 8-10 second voltage bi-ramps from -70 mV to 10-40 mV. PICs were observed in 72% of EGFP neurons (34/47) in all laminae examined. PICs were activated at about -58.4±8 mV (n=28) with amplitude of -66.3±50 pA and conductance of 2.3±2 nS. PICs were also observed in recordings with TTX (0.5-1 uM, n=16). PICs could be blocked by nifedipine (10-30 uM, n=4) or riluzole (10-50 uM, n=5) with or without TTX. PICs with two components (n=4) which activated at different voltages on the rising phase of the voltage ramp were observed. These results suggested that PICs could be mediated by L-type calcium and/or persistent sodium currents and activated at different voltages in the EGFP+ neurons. 5-HT modulation of PICs usually consisted of a hyperpolarization of activation voltage and an increment of the amplitude of the PICs (10/14). 166.10/AA29 - Innervation of the extended amygdala by the paraventricular nucleus of the thalamus *G. J. KIROUAC1, M. P. PARSONS2, S. LI1; 1Department of Oral Biology, University of Manitoba, Winnipeg, MB, CANADA, 2Division of Basic Medical Sciences, Memorial University, St. John's, NF, CANADA. The paraventricular nucleus of the thalamus (PVT) is part of a group of midline thalamic nuclei implicated in arousal. Previous tracing studies indicate that the PVT projects to the nucleus accumbens, medial prefrontal cortex, subiculum and several nuclei of the amygdala and bed nucleus of the stria terminalis (BST) associated with the extended amygdala (EA). This study examined the connections between the PVT and its targets with a special focus on the EA. The retrograde tracer cholera toxin B (CTb) was iontophoretically injected in different regions of the forebrain. The posterior PVT contained large numbers of CTb-labeled neurons following injections in the lateral BST and central nucleus of the amygdala (CeA). Injections of CTb in the basolateral nucleus of the amygdala resulted in moderate number of labeled neurons in the posterior PVT whereas injections in the medial BST or the medial nucleus of the amygdala produce much fewer numbers of labeled neurons. Injections of the anterograde tracer biotin dextran amine (BDA) in the posterior PVT produced heavy fiber labeling that extended from the medial shell of the nucleus accumbens to the lateral BST, interstitial nucleus of anterior commissure (IPAC) to the CeA. The medial BST, medial nucleus of the amygdala and the basolateral nucleus of the amygdala contain much less extensive BDA fiber labeling. In addition, double labeling experiments looking at the distribution of BDA fibers with tyrosine hydroxylase labeling revealed that the PVT preferentially innervates regions of very dense tyrosine hydroxylase fibers in the shell of the nucleus accumbens and central division of the EA. We conclude that the posterior PVT represents the only nuclei of the midline and intralaminar group to innervate the central EA. Furthermore, the efferents from the posterior PVT innervate the tyrosine hydroxylase rich regions of the shell of the nucleus accumbens and the central EA. The PVT may be part of an arousal system related to viscerolimbic functions. 168.29/CC17 - Cultured cortical neurons from embryonic TgCRND8 Mice display aberrant calcium regulation and increased cell death *K. L. OLSON1, J. SCHAPANSKY1, P. SALINS2, H. YANG2, F. M. AMARA2, G. W. GLAZNER1; 1Pharmacology & Therapeutics, St Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA, 2Biochemistry and Medical Genetics, St. Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA. The TgCRND8 transgenic mouse model of Alzheimer disease (AD), which contains the double mutant Swedish and Indiana (KM670/671NL + V717F) form of the most common APP isoform, APP695, displays elevated levels of cortical Ab and cognitive deficits reminiscent of AD by 3 months of age and cortical amyloid plaques by 5 months. Thus, the TgCRND8 mouse is a highly relevant animal model of AD. However, to our knowledge, the characteristics of this model in primary embryonic cortical neuron culture have not been reported. Although increased Ab1-42 and mutant APP are not expressed in TgCRND8 mice until weeks after birth, cultured neurons often develop at a more rapid rate. Thus, we hypothesized that cultured embryonic TgCRND8 cortical neurons would express mutant APP and elevated Ab1-42, and display cellular pathology. Homozygous WT females were crossed with heterozygous TgCRND8 males and embryos removed for culturing at E16. Presence of TgCRND8 was determined by PCR for each embryo prior to dissection and establishment of cultures, and each litter was separated into TgCRND8 and WT littermates for in vitro testing. Initial survival and neurite outgrowth (1-3 days) was not different between TgCRND8 and WT cultures, and APP and Ab1-42 levels were similar at that time. However, by 8 days in culture, TgCRND8 neurons expressed significantly higher levels of APP than WT cultures, and after 14 days TgCRND8 neurons generated significantly greater levels of Ab1-42 compared to WT cultures. From 12 days in culture, TgCRND8 cultures demonstrated significantly increased background death than WT cells, and elevated concentrations of intracellular [Ca+2]. In addition, the acute increase in intracellular [Ca+2] generated by exposure to 100 mM glutamate was significantly greater in TgCRND8 vs. WT cortical cultures. It is important to note that no pathological differences were seen between TgCRND8 and WT neurons prior to increased APP expression. Our data indicate that cultured embryonic TgCRND8 cortical neurons express the mutant APP and generate increased levels of Ab1-42, and that this induces oxidative stress, loss of calcium homeostasis, and neuronal death. TgCRND8 mice may be a valuable model for in vitro studies of the effects of endogenously produced Ab1-42. 172.20/FF24 - Investigation of the relationship between nuclear factor kappa B (NF-kB) activity and long-term potentiation (LTP) in hippocampal slices of an Alzheimer’s disease mouse model (M146V) *B. C. ALBENSI1, G. ODERO1, K. COLLISTER1, F. AMARA2, G. GLAZNER1, K. OIKAWA1; 1Pharmacology & Therapeutics, St Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA, 2Biochemistry & Medical Genetics, St. Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA. Increasing reports suggest that the disruption of calcium homeostasis plays a decisive role in the pathophysiological process of Alzheimer’s disease (AD). In addition, cells expressing mutant presenilin-1 (PS1), a protein involved in early-onset AD, exhibit an aberrant pattern of NF-kB activation following excitotoxic insults involving high intracellular calcium concentration. NF-kB mediated signaling is complex and plays a critical role in many biological processes including autoimmune diseases, cancer, inflammatory responses, and in cellular neuroprotection. Investigators have also reported that NF-kB is activated during the induction of LTP and may be a requirement for synaptic plasticity and memory, which has up to now not been fully realized. In an attempt to investigate NF-kB in the pathogenesis of AD, we used 1, 3, 6, and 12 month-old transgenic mice M146V and C57BL/6 (background control). We found that LTP induced by 3 trains of 100 Hz high frequency stimulation was differentially altered in an age-dependent manner in M146V as compared to control mice. Additionally, we performed ELISA-based and electromobility shift transcription assays to evaluate the modulation of NF-kB activity in hippocampal brain slices of M146V and C57BL/6 mice at 1 hr after LTP induction. Interestingly, we found that NF-kB activation levels were increased under basal conditions in the M146V slices compared with C57BL/6. However, upon 100 Hz stimulation, NF-kB activation levels increased in C57BL/6 slices, but not in M146V slices. In addition, we performed Western blotting to evaluate expression level of p50 NF-kB subunit and calbindin D28K, which is one of the target genes of NF-kB, in hippocampus tissue homogenate from M146V versus C57BL/6. We found that p50 NF-kB expression was decreased in dentate and that calbindin D28k expression was decreased in CA3 in M146V mice. These data present an important linkage between PS1 mutations and the aberration of NF-kB activation. 185.8/LL102 - Quercetin-3-glucoside and kaempferol activate the NAD(P)H:quinone oxidoreductase (NQO1) antioxidant response element (ARE), increase NQO1 activity, and protect cortical neuronal cultures from oxidative stress-induced death *C. A. NOONAN1,2, J. A. GRANT1,3, C. B. SUPNET1,2, M. MAYNE1,2,3; 1Institute for Nutrisciences and Health, Natl Research Council, Charlottetown, PE, CANADA, 2Biomedical Sciences, University of Prince Edward Island, Charlottetown, PE, CANADA, 3Pharmacology and Therapeutics, University of Manitoba, Winnipeg, PE, CANADA. Elevated concentrations of reactive oxygen species (ROS) are observed in brain post-mortem specimens from neurodegenerative patients, suggesting a disruption in clearance of oxidized products. Excess ROS threaten neuron viability by causing lipid peroxidation, and DNA and protein oxidation. In addition to possessing direct antioxidant properties, quercetin, a well known flavonoid, has been shown to activate signalling pathway(s) that engage a gene enhancer known as the antioxidant response element (ARE). ARE induction results in the production of phase 2 detoxification and antioxidant enzymes, including NQO1 (NAD(P)H:quinone oxidoreductase), which play an important role in the fight against oxidative stress and harmful xenobiotics. A core ARE sequence is conserved across proteins and between mice and humans. Our lab confirmed that quercetin does activate an ARE, but at concentrations that are toxic to murine neuronal cultures. In further studies, we have found that the flavonoids quercetin-3-glucoside and kaempferol also induce phase 2 enzymes & antioxidant enzymes transcription. A 48 hr incubation of U87 cells with quercetin-3-glucoside (Q3G) or kaempferol significantly elevated ARE induction and the activity of NQO1 through a dicumarol-inhibitable reaction. Finally, we have found that 48-hr pre-treatment with Q3G or kaempferol significantly protected C57/B6 primary cortical neurons from hydrogen peroxide-induced cell death (LD50 425 mM at 24 hr). We are currently determining the extent to which flavonoid neuroprotection is mediated via astrocytes. Chair L. M. JORDAN; Dept Physiology, Univ Manitoba, Winnipeg, MB, CANADA. Restoration of the ability to walk after injury or disease is a high priority for regeneration and transplantation approaches, and knowledge of the neurons involved in the initiation and control of locomotion is required if such efforts are to succeed. The neural control of locomotion has been a central theme of motor control for many decades, but our knowledge of the neurons involved is still very limited. Recent advances in neural development and molecular genetics have led to new approaches to the identification of these neurons. We will review recent data leading to the identification, properties and function of locomotor neurons of the brainstem and spinal cord, assess how far the field has progressed, and gain some insight into work that remains to be done to define the ensemble of neurons controlling locomotion. Dr. Goulding will review genetic approaches to defining classes of spinal cord interneurons, including developmental studies that have identified four classes (V0, V1, V2, V3) of embryonic interneurons in the ventral spinal cord, transcription factors that define the classes of interneurons, and their role in locomotion determined using targeted deletions of specific classes of interneurons. Dr. Ziskind-Conhaim will discuss synaptic and cellular mechanisms that underlie locomotion rhythms in specific classes of GFP-expressing interneurons, functionally identified and visually targeted for electrophysiological and morphological studies, and present data elucidating their probable roles in the locomotor CPG. Dr. Ole Kiehn will describe a genetic loss-of-function related to a distinct abnormal behavioral phenotype, the rabbit-like hopping gait exhibited in the isolated spinal cords of mice with a targeted deletion of the axon guidance molecules EphA4 and ephrinB3, and studies testing the hypothesis that a group of EphA4 neurons are excitatory interneurons in the locomotor CPG and are involved in producing the ipsilateral drive during locomotion. Dr. Kiehn will also discuss the neural systems coordinating left and right motor activities. Dr. Jordan will describe new data on the use of c-fos/EGFP transgenic mice to visually identify locomotor interneurons, and on identified reticulospinal neurons involved in the initiation of locomotion. The presentations in this symposium will show how classical electrophysiological and anatomical techniques can be combined with state of the art genetic techniques to gain unprecedented insights into CNS function. 105.5 - Command Systems in the Ensemble of Neurons Controlling Locomotion L. M. JORDAN; Physiology, University of Manitoba, Winnipeg, MB, CANADA. 252.3/U20 - Influence of the lumbar cord on cervical locomotor-like activity evoked by electrical stimulation of the neonatal rat brainstem in vitro *K. C. COWLEY, E. ZAPOROZHETS, B. J. SCHMIDT; Dept Physiol, Univ Manitoba, Winnipeg, MB, CANADA. Although the locomotor network is distributed throughout the spinal cord, certain regions of the cord are likely to have greater capacity for rhythm generation and limb coordination. In previous work, we examined the influence of propriospinal relays in the cervicothoracic region on the ability of brainstem electrical stimulation (BES) to induce locomotor-like activity in the lumbar cord (Zaporozhets et al. J. Physiol. 572: 2006). In the present study we examine whether the lumbar region has an influence on locomotor-like rhythm in the cervical cord, evoked by BES. Intact in vitro brainstem-spinal cord preparations were isolated from neonatal rats (days 0-4). Ventral root discharge was monitored at the cervical, thoracic and lumbar levels. Complete cord transection at the lower thoracic level (e.g. T10), or suppression of lumbar rhythmic activity by removal of calcium from the ACSF bathing the lumbar cord, terminated locomotor-like activity in the cervical region. Mid-sagittal section of the cord from T12 to L5, inclusive, abolished lumbar locomotor-like activity in response to BES, and also terminated cervical rhythm generation. In contrast, cervical rhythmic activity induced by bath application of 5HT/NMDA persisted after lower thoracic cord transection or mid-sagittal section (T12-L5), although ventral root discharge in the cervical and thoracic regions was slower and bilaterally synchronous under these conditions. In other experiments, electrical stimulation was applied at the T2 cord level in preparations with complete transections at the T1/2 level; thoracic ventral root discharge, phase-related to lumbar locomotor-like activity, was elicited. After transection at the T10 level, electrical stimulation at T2 failed to activate rhythmic discharge in the thoracic cord, however 5HT/NMDA was still capable of inducing rhythmic activity in the isolated thoracic cord. These results suggest that although the cervical and thoracic regions have inherent rhythmogenic capacity independent of the lumbar network, the development of locomotor-like activity in the cervical and thoracic cord in response to BES depends on the generation of rhythmic activity in the lumbar region. 252.20/V15 - 5-HT7 receptor blockade disrupts voluntary locomotion: studies in adult wild-type and 5-HT7 receptor knock-out mice J. LIU1, T. AKAY2, K. G. PEARSON2, P. B. HEDLUND3, *L. M. JORDAN1; 1Dept Physiology, Univ Manitoba, Winnipeg, MB, CANADA, 2Dept Physiology, Univ Alberta, Edmonton, AB, CANADA, 3Dept Molecular Biology, Scripps Research Institute, La Jolla, CA. We have recently shown that a descending pathway from the medulla containing 5-hydroxytryptamine (5-HT) leads to locomotor-like activity in rodents. The initiation of locomotion by this pathway requires activation of 5-HT7 receptors (Liu & Jordan, J Neurophysiol 94: 1392-1404, 2005). Hedlund et al (PNAS 100:1375-1380, 2003) generated a mouse strain with a targeted disruption of the 5-HT7 receptor gene (5-HT7-/- mice). We previously used this mouse strain to show that locomotor-like activity produced by 5-HT applied to the isolated neonatal spinal cord requires functional 5-HT7 receptors (Liu et al, Soc. for Neurosci. 516.6, 2005). In the current study we tested the hypothesis that 5-HT7 receptors are involved in the control of voluntary locomotion in adult mice. A method developed in Edmonton for monitoring overground voluntary locomotion in adult mice during direct subdural application of drugs was used. The animals were anesthetized with isoflurane, and the analgesic buprenorphine (0.15mg/kg) was injected SC. A laminectomy was performed to expose the L2 and L3 spinal segments, and the dura matter over these segments was removed. Saline or SB-269970 was then applied to the exposed cord for 30 minutes. Anaesthesia was then terminated, the saline or drug was removed, and the animal was placed on a confined walkway, where it recovered from the isoflurane anesthesia. Spontaneous locomotion usually commenced within a few minutes. SB269970, a specific 5-HT7 receptor antagonist, consistently altered locomotion in wild type animals, producing over-extension of both hindlimbs such that the stance phase often persisted beyond the point where weight support could be maintained. This resulted in slowed locomotion and “dragging” of the hindlimbs for a period of 30 to 90 minutes, until recovery occurred. In order to show that the effect of SB-269970 was due to an action on 5-HT7 receptors, mice lacking 5-HT7 receptors were also treated with SB-269970. In the vast majority of these cases, the effects were far less severe than in SB-269970 treated wild type animals. These results are consistent with our hypothesis that activation of 5-HT7 receptors plays an important role in the control of locomotion in adult mice. 271.3/KK4 - Calreticulin protein levels are increased in in vitro and in vivo models of Alzheimer’s disease *J. A. GRANT1,2, C. B. SUPNET1,3, C. A. NOONAN1,3, M. MAYNE1,2,3; 1Institute for Nutrisciences and Health, National Research Council, Charlottetown, PE, CANADA, 2Pharmacology and Therapeutics, University of Manitoba, Winnipeg, MB, CANADA, 3Biomedical Sciences, University of Prince Edward Island, Charlottetown, PE, CANADA. Several studies have indicated that increased oxidative stress and disruption of intracellular calcium homeostasis are associated with the neuronal dysfunction and neurotoxicity seen in Alzheimer’s disease (AD). Calreticulin (CRT) is a 46 Kda calcium-binding protein that is found within the ER, which functions to regulate calcium levels and to chaperone proteins within the endoplasmic reticulum (ER). CRT expression is induced by a variety of biological stressors, and induction of CRT expression increases the free calcium concentration within the lumen of the ER, resulting in increased calcium responses from ER stores and increased cytosolic calcium concentrations. CRT also plays an important role in the induction of apoptosis in cells exposed to ER stress. Here, we report an increase in CRT protein levels in SH-SY5Y human neuroblastoma cells exposed to oxidative stress and Amyloid beta (Ab) protein. Briefly, SH-SY5Y cells were treated with H2O2 (25-50mM), Ab25-35 (5-20mM) and Ab1-42 (5-20mM) for 24 hours, and CRT protein levels were determined via western blot analysis. A significant and dose-dependant increase in CRT levels was seen in all three treatments as compared to untreated controls. In addition, CRT protein levels were increased in the cortex of transgenic (Tg)CRND8 mice, which are a mouse model of AD. A significant increase in cortical CRT levels was seen in 12 and 16 week-old TgCRND8 mice as compared to aged-matched littermate controls. In conclusion, we have reported an increase in CRT protein levels in SH-SH5Y neuroblastoma cells exposed to AD-related stressors and in the cortex of TgCRND8 mice as compared to Non-Tg CRND8 littermate controls. We are currently determining the role of CRT in the disruption of calcium homeostasis and subsequent induction of apoptosis seen in AD using siRNA technology, as well as determining CRT levels in multiple age groups and brain regions of TgCRND8 mice as compared to their controls. 343.5/P5 - Astrocytes influence N-methyl-D-asparate-evoked release of adenosine from neurons *C. R. ZAMZOW, W. XIONG, F. E. PARKINSON; Dept Pharmacol & Therapeut, Univ of Manitoba, Winnipeg, MB, CANADA. The purine adenosine (ADO) has protective effects in stroke models and NMDA receptor activation can increase ADO in the CNS. Previous research from our laboratory has shown that cultured cortical neurons release ADO per se through equilibrative nucleoside transporters (ENT) whereas astrocytes produce ADO through an extracellular pathway. As well, astrocytes have been shown to express more ecto-5’ nucleotidase (e-N), the enzyme responsible for extracellular dephosphorylation of AMP to ADO. We hypothesize that the presence of astrocytes alters NMDA-evoked purine release from neurons. All primary forebrain cortical neurons were harvested from E17 Sprague Dawley rat whereas astrocytes cultures were harvested from P1 rat forebrains. Neurons were pre-loaded with [3H]adenine to radiolabel adenine nucleotides then treated with NMDA (100 mM) for 30 minutes at 37°C after which supernatant samples were collected and analyzed for ADO, inosine (INO), and hypoxanthine (HX). To determine the role of astrocytes, neurons were: treated with astrocyte-condition media, grown on a layer of astrocytes (co-cultures), or had astrocytes on transwells added during treatment. Data were analyzed with one-way ANOVA and Tukey’s post-hoc tests. Control ADO and INO release from neurons was 10.1 ± 1.7 and 30.8 ± 6.1 pmol/mg protein, respectively. NMDA significantly increased ADO and INO release from neurons to 27.2 ± 4.6 and 98.5 ± 20.3 pmol/mg protein. The ENT inhibitor dipyridamole (DPR) prevented NMDA-evoked ADO release whereas the e-N inhibitor alpha, beta-methylene ADP (AOPCP) had no significant effect, indicating an intracellular source of ADO and INO release from neurons. Astrocytes stimulated with NMDA had no significant increases in ADO or INO release. Astrocyte-conditioned media did not alter ADO release from neurons following NMDA. In co-cultures, control ADO and INO release was decreased whereas HX increased, relative to neuron cultures, and AOPCP significantly inhibited release of ADO. AOPCP also inhibited NMDA-evoked ADO release when astrocytes on transwells were added to the neurons during treatment whereas DPR had no effect. In summary, NMDA-treated neurons release ADO and INO through ENT. However, in the presence of astrocytes, the extracellular pathway for ADO release becomes significant. Therefore, NMDA-evoked ADO release from neurons is influenced by astrocytes. 377.13/MM46 - Visualization and correlations involving the deposition of amyloid beta plaques in the TgCRND8 mouse model of Alzheimer's disease *K. A. COLLISTER1,2, G. ODERO2, M. MARTIN3, B. C. ALBENSI1; 1Pharmacology & Therapeutics, University of Manitoba, Winnipeg, MB, CANADA, 2Division of Neurodegenerative Disorders, St. Boniface General Hospital Research Centre, Winnipeg, MB, CANADA, 3Physics, University of Winnipeg, Winnipeg, MB, CANADA. Neuropathology resulting from Alzheimer’s disease (AD) includes neurodegenerative changes such as accelerated amyloid beta (Ab) plaque deposition and neurofibrillary tangles. Other human hallmarks of AD consist of gradual cognitive decline, a decreased ability to understand and communicate, and an inability to form new memories. This study investigates the development of MRI methods for the earlier detection of AD and an exploration into mechanisms related to AD memory impairment. Fourteen-month old transgenic CRND8 mice and their littermates (background controls) brains were scanned with MRI to directly visualize amyloid beta plaque deposition. We used in vivo and ex vivo T2-weighted imaging to visualize plaques, as well as diffusion-weighted imaging (DWI) to quantify the effects of plaque deposition on the diffusion of water. From the diffusion-weighted raw images, apparent diffusion coefficient (ADC) maps were produced. Post-mortem, the mouse brains were sliced and immunohistochemically stained with BTA fluorescent tags to visualize plaques, as well as with fluorescent tags for NF-kB and CREB, transcription factors potentially involved in synaptic plasticity and memory formation. The T2-weighted images and immunohistochemical staining showed the deposition of plaques to be abundant in the CRND8 hippocampus. The ADC values in the hippocampus were altered in the transgenic mice suggesting that ADC changes in the hippocampus may serve as an early indicator of amyloid plaque deposition in Alzheimer’s disease. 428.12/D15 - Hippocalcin, a possible Ca2+-dependent regulator for microsomal mono-oxygenase system, interacts with amyloid b1-42 *S. M. DOMBROWSKI1, B. ALBENSI2, G. ODERO3, K. COLLISTER4, K. OIKAWA3; 1Dept Neurosurgery, Cleveland Clinic Fndn, Cleveland, OH, 2Dept of Pharmacology and Therapeutics, Univ of Manitoba, Manitoba, MB, CANADA, 3Dept of Pharmacology and Therapeutics, St. Boniface Research Centre, Winnipeg, MB, CANADA, 4Dept of Pharmacology and Therapeutics, University of Manitoba, Manitoba, MB, CANADA. Hippocalcin is a high-affinity calcium-binding protein restricted to the brain, most abundant in pyramidal cells of the hippocampal CA1 region. Evidence demonstrates that hippocalcin rapidly translocates from the cytosol to the perinuclear compartment responding to increases in intracellular Ca2+ concentration. Kobayashi et al. reported that hippocalcin may be essential for long-term plasticity including spatial and associative memory. More recently, it has been indicated that hippocalcin plays a critical Ca2+-sensing role in NMDAR-mediated hippocampal long-term depression (LTD). Although it is still hard to assign the physiological function of hippocalcin, we discovered that hippocalcin interacts with microsomal cytochrome b5 (Cyb5) in a Ca2+-dependent manner (J Biol Chem. 2004; 279(15):15142-52). These findings suggested that hippocalcin may link Ca2+-signaling to the machinery of microsomal monooxygenase (MMO) complex composed of Cyb5, cytochrome P450, and some reductases in the endoplasmic reticulum. Recently we obtained evidence that VILIP-3, which has 94% identity with hippocalcin, can also interact with Cyb5-Reductase (b5R). Furthermore, Ca2+- bound VILIP-3 enhanced the interaction between Cyb5 and b5R inversely, Ca2+-unbound VILIP-3 attenuated the binding force between them. At this stage, we hypothesize that when intracellular Ca2+ concentrations reach a high level, Ca2+-bound hippocalcin, as well as VILIP-3, rapidly translocates from cytosol to ER, and enhances the coupling between Cyb5 and b5R in hippocampal CA1 pyramidal neuron. In the amino acid sequence alignment, we revealed that amyloidgenic Aß42 and Cyb5 both have the AIxxxxV sequence. This finding raises the possibility that Aß42 can bind to hippocalcin. To test this possibility, we performed an Aß42 pull-down assay. As estimated, hippocalcin was captured by the aggregated Aß42. Taken together, in the earliest stage of AD, Aß42 may perturb the signaling pathway involving hippocalcin by competition with Cyb5 in the ER for Ca2+-dependent and orderly interaction between hippocalcin and Cyb5 440.20/K13 - Differences in the temporal integration limits of stereovision in perception and action *K. R. WILSON1, M. A. ATKINS2, P. M. PEARSON3, B. TIMNEY2, J. J. MAROTTA1; 1Dept of Psychology, Perception and Action Lab, Univ Manitoba, Winnipeg, MB, CANADA, 2Dept of Psychology, Univ Western Ontario, London, ON, CANADA, 3Dept of Psychology, Univ Winnipeg & Univ Manitoba, Winnipeg, MB, CANADA. Objectives. Alternating monocular views can be integrated to achieve stereovision, even when there is an interocular delay of up to 100ms. While these studies have typically been carried out in the perceptual domain, it has been well established that binocular vision is of critical importance for the programming and control of visually guided grasping. We were interested in comparing the performance of the “perception” and “action” visual cortical streams to different amounts of interocular delay. Methods. Stereopsis and reach kinematics were assessed under binocular, monocular, and alternating monocular viewing conditions with the aid of LCD goggles. In the alternating monocular viewing conditions, the left and right shutters were alternately opened for a period of 50 ms each with interocular delays of 15, 35, or 60 ms. To assess the temporal integration limits for stereopsis, the percentage of trials in which participants could correctly determine whether the disparity (6 min) in a random dot stereogram was crossed or uncrossed was measured. To assess whether alternating monocular input can be integrated to efficiently guide reaching and grasping, maximum velocity and grip aperture was measured while participants reached out and picked up objects placed at one of three different distances along their midline. Results. In the perceptual task, stereopsis was achievable even with a 60 ms interocular delay. In contrast, even the shortest interocular delay impaired reaching and grasping movements relative to the monocular and binocular viewing conditions. Binocular viewing produced faster limb velocities and significantly smaller grip apertures, a measure of grasp efficiency, than all interocular delay conditions. Conclusions. The results from the stereopsis task confirm that the perceptual experience of depth is tolerant to interocular delays. On the other hand, the benefit of binocular viewing in visually guided reaching appears to be degraded by even short interocular delays, suggesting asymmetrical temporal integration limits in the two streams of visual processing. This fits well with the theory that the vision-for-action system is one that deals with moment-to-moment information and would have no use for “persistent” representations of depth. 440.21/K14 - A new window into the interactions between perception and action L. A. BAUGH, *J. J. MAROTTA; Dept. Psychology, Perception and Action Lab, Univ of Manitoba, Winnipeg, MB, CANADA. Objectives. Traditionally, viewing window paradigms have been solely used to evaluate perceptually based features useful in object identification. In the simplest form, participants are presented with a degraded picture of an object on a computer monitor and are asked to identify the object as quickly as possible. A small, user controlled area (the "window") displays the underlying image with normal clarity. Despite their traditional role, viewing window tasks inherently require visuomotor processing, which can be manipulated to illuminate the interactions between the “perception” and “action” based cortical visual streams. Methods. Twelve young adults were recruited from the University of Manitoba participant pool (6M, 6F; Age Range 17 - 22 years old; Mean Age = 18.5) and identified 43 common objects presented using the viewing window task. Participants were asked to identify the presented object as quickly as possible. The viewing window was controlled by the user via a touchscreen, ensuring a 1-1 correspondence between motor acts and window movements. Response times (ms) and movement of the window (represented by a series of x-axis and y-axis coordinates) were recorded. Results. Response times were separated into two distinct categories: Time spent before moving the focus-window (pre-movementRT) and time spent moving the focus window (movementRT). A correlational analysis was performed ensuring pre-movementRT was not significantly correlated with movementRT (r = .05), which would have been indicative of diagnostically useful information being presented outside of the viewing window. Scanning pattern was analyzed by separating the image into four equal sized quadrants (top-left, top-right, bottom-left, and bottom-right) and then performing a repeated measures ANOVA on the percentage of time participants spent in each. A significant main-effect of quadrant was found (F = 32.6, p < .001), demonstrating the ability of this procedure to identify vertical and horizontal asymmetries in visuomotor scan patterns elicited by the presented objects. Further analysis of visuomotor scan paths at a higher resolution revealed the location of features for each object that received the most attention. Conclusions. This research provides a strong experimental basis for studies examining spatial and perceptual distortion effects in object recognition, the effects of aging on visuomotor control, and disorders of spatial representation using the viewing window paradigm. 448.11/U2 - Effect of selected spinal cord lesions on locomotor-like activity evoked by electrical stimulation of the neonatal rat brainstem in vitro E. ZAPOROZHETS, K. C. COWLEY, *B. J. SCHMIDT; Dept Physiology, Univ Manitoba, Winnipeg, MB, CANADA. We recently showed that a descending propriospinal system helps mediate brainstem activation of locomotor networks in the in vitro neonatal rat brainstem-spinal cord preparation (Zaporozhets et al. J. Physiol. 572: 2006). In the present work selected cord lesions were made, using the same preparation (days 0-4), in an effort to determine the location of critical components of the bulbospinal locomotor command pathway and rhythm generating networks targeted by these pathways. Brainstem electrical stimulation (BES) was used to induce locomotor-like activity that was monitored using lumbar ventral root recordings. In preparations with staggered contralateral hemisections (e.g. right T2 and left T10), which disrupted all direct brainstem projections to the lumbar cord, BES still induced locomotor-like activity, suggesting that cross-connected descending propriospinal neurons alone may be sufficient for the propagation of the locomotor command signal. Bilaterally coordinated locomotor-like activity in the lumbar region was elicited in preparations with thoracic cord lesions that spared only the ventrolateral fasciculus on one side. Mid-sagittal lesions of varying length were made at different locations between C1 and the conus. These experiments showed that commissural projections were redundantly distributed throughout the cord such that cross connections at no particular segmental level were essential for bilaterally coordinated locomotor-like activity in response to BES. No mid-sagittal lesion, nor combination of mid-sagittal lesions, was capable of uncoupling alternating left-right rhythmic discharge in the lumbar cord. However, preservation of rhythmic discharge evoked by BES in the presence of mid-sagittal lesions required that at least some commissural connections remain intact between T12 and L2. Mid-sagittal lesions from T12 through L2 inclusive abolished rhythmic activity induced by BES, in contrast to the lack of effect similar lesions had on locomotor-like activity induced by bath application of neurochemicals (Cowley et al. J. Neurophysiol. 77: 1997). In combination, these observations suggest that lumbar portions of the spinal locomotor network can be activated unilaterally by a bulbospinal signal transmitted in the VLF, but at least some commissural components must be preserved in the thoracolumbar region in order for either side of the cord to generate rhythmic activity. 465.21/II24 - Functional and anatomical connection between the paraventricular nucleus of the thalamus and dopamine fibers in the nucleus accumbens *M. P. PARSONS1, S. LI2, G. J. KIROUAC2; 1Div. Bas. Med. Sci., Memorial Univ, St. John's, NF, CANADA, 2Department of Oral Biology, University of Manitoba, Winnipeg, MB, CANADA. The shell of the nucleus accumbens (NacSh) receives a dense innervation from dopamine neurons in the ventral tegmental area and glutamate neurons in the paraventricular nucleus of the thalamus (PVT). The present study examined the effects of brief electrical stimulation of the PVT on electrochemically detected dopamine release in the NacSh of urethane anesthetized rats. Transient stimulation of the PVT (40 Hz, 1.0 ms, 400 µA, 5 s) resulted in a brief increase in electrochemically detected dopamine in the NacSh. Inhibition of dopaminergic neurons in the ventral tegmental area using lidocaine (4%, 500 nl) or intravenous apomorphine (0.15 mg/kg; i.v.) decreased the resting voltammetric signal but had no effect on PVT-evoked responses. Blocking of ionotropic receptors in the NacSh with local administration of kynurenic acid attenuated the PVT-evoked responses in the NacSh. Anterograde tracing with biotinylated dextran amine demonstrates that PVT afferents target regions of very dense tyrosine hydroxylase fiber staining in the NacSh. Consistent with the projection pattern of the PVT to the NacSh, stimulation of the PVT evoked the largest oxidation current changes in the NacSh whereas small or little changes were elicited in other areas of the striatum. This study suggests that glutamate release from PVT terminals can act on ionotropic glutamate receptors in the NacSh to induce dopamine efflux. Modulation of dopamine levels in the NacSh by the PVT may be linked to arousal-induced increases in dopamine tone and could be involved in the facilitation of specific behavioral patterns associated with arousal or stressful situations. 467.21/KK8 - A combination of physical and cognitive activity prevents cognitive deterioration in a rat model of vascular dementia *D. R. CORBETT1, C. HARLEY2, P. MOODY-CORBETT1, G. CHERNENKO1, S. GRANTER-BUTTON1, K. MCKAY1, A. ORSBORN1, R. BUIST3, J. PEELING3; 1Fac Med, Memorial Univ Newfoundland, St John's, NF, CANADA, 2Dept. of Psychology, Memorial Univ Newfoundland, St John's, NF, CANADA, 3Pharmacology and Therapeutics, University of Manitoba, Winnipeg, MB, CANADA. While human data suggest both cognitive and physical activity may be beneficial in reducing the risk of dementia, experimental data examining both parameters individually, and in combination, in an animal model of dementia have not been investigated. We reduced cerebral blood flow by permanent carotid artery occlusion in middle aged male rats. Rats were then assigned to normal or high fat, high sugar diets. Rats given normal diets were given access to running wheels (physical activity), or complex housing in Hebb-Williams mazes (cognitive activity), or both. All rats had social housing. Another sham and a carotid occlusion group were housed in an enriched environment. Rats were tested in the Morris water maze 1, 2, 4 or 6 mo after carotid occlusion. The first two tests used an insoluble Morris water maze problem to familiarize the rats with the task requirements. All groups (shams and carotid occluded) showed reduced swim latencies on the insoluble maze problem. Carotid occlusion prevented spatial learning at 4 mo, while sham operated groups and carotid occluded rats receiving combined physical and cognitive treatments showed significant spatial learning and were not different from each other. At 6 mo spatial learning was again significantly faster in the sham groups and in the combined physical and cognitive treatment carotid occluded group. All groups, however, showed some spatial memory in the probe test at 6 mo. The percentage of rats exhibiting significant memory performance was highest in the sham and combined physical and cognitive therapy groups. These data demonstrate that combined physical and cognitive therapy prevent the cognitive deficits normally seen with permanent carotid artery occlusion. Either therapy alone was intermediate or non-significant in effect. With physical and cognitive therapy combined, rats with permanent carotid occlusion were indistinguishable in the Morris water maze from sham operated rats maintained in an enriched environment. An investigation of the underlying mechanisms of the beneficial action of this combined physical and cognitive intervention is under investigation. 492.1/PP76 - Prospective removal of cardiac-induced signal contaminants in fMRI time series *N. H. NEUFELD1, S. MEHTA2, L. BOLINGER3, T. J. GRABOWSKI2,4, M. C. MCINTYRE1; 1Psychology, Univ Winnipeg, winnipeg, MB, CANADA, 2Neurology, Univ Iowa, Iowa City, IA, 3MR Research and Development, National Research Council Canada, Institute for Biodiagnostics, Winnipeg, MB, CANADA, 4Radiology, Univ Iowa, Iowa City, IA. OBJECTIVE: Cardiac pulsation causes magnetic resonance signal changes that act as noise in functional magnetic resonance imaging (fMRI) time series, reducing the sensitivity of data analysis. Temporal autocorrelation of residual terms may also be introduced, threatening the validity of statistical inference. The shift from block design to event-related design, with its lower sensitivity, emphasizes the need to address cardiac-induced noise. This study compares three approaches to modeling cardiac-induced noise as an effect of no interest within the general linear model. METHODS: Using the I/OWA system (Smyser et al., 2001), anatomical and functional images were obtained for three subjects on a 1.5 Tesla GE LX CV/i scanner. The image acquisition and cardiac channels were filtered to gather timing information. It was determined at what phase of the cardiac cycle each image was acquired. A voxel-wise deconvolution was then performed using three basis sets (Meizin et al., 2000): Fourier deconvolution using a truncated Fourier series with the relative position within the R-R interval of the cardiac cycle as the phase (Glover et al., 2000), bin deconvolution using the relative time within the cardiac phase indicated by the R-wave, and bin deconvolution using the absolute time from the last R-wave. RESULTS: Adjusted R2 images showed the three approaches similarly modeled cardiac-related signal changes in brain blood vessels across three subjects. Cardiac-induced noise was modeled in grey matter and showed spatial heterogeneity within and between subjects. Grey matter close to blood vessels and pulsation-susceptible tissues were better modeled by all approaches than grey matter regions more distant from said regions. CONCLUSIONS: The three approaches all accounted for signal variance to an equal or greater extent in all examined regions than a single cosine function. Spatial heterogeneity in the amount of cardiac-induced signal variance between and within subjects underscores the importance of individual differences and provides further rationale for prospective modeling. Any one of the approaches may be used in fMRI statistical packages or real-time analysis to prospectively remove cardiac-induced signal changes. 506.10 - Face-name recognition following selective medial temporal resections *D. H. ZALD1, S. D. SMITH2; 1Dept Psychol, Vanderbilt Univ, Nashville, TN, 2Dept Psychol, University of Winnipeg, Winnipeg, MB, CANADA. Substantial research indicates that the medial temporal lobes are critical for the formation of paired-associate representations. Associating face-name pairs appears to be particularly challenging in that even healthy controls often require repetition to acquire such pairings. Neuroimaging studies have observed activations in the right, left or bilateral medial temporal lobe during face-name memory tasks. In order to better understand what aspects of face-name associative learning are affected by the right vs. left medial temporal lobe, we examined face-name recognition in patients who had undergone selective unilateral resection of the amygdala and anterior hippocampus for intractable epilepsy. Thirteen adult patients (7 left-hemisphere, 6 right-hemisphere) and 9 age and education matched controls were shown photographs of male faces, each paired with a common male name. After viewing all the face-name pairs, participants were shown face-name pairings, and had to indicate whether they had seen the face-name pairing during the learning phase. The three groups were similar in their ability to reject new faces paired with new names (all > 87% accuracy), indicating that patients with unilateral medial temporal lobe damage have no difficulty identifying which items they have no familiarity with. In contrast, there was a significant difference in the ability to recognize previously seen face-name pairings, with the left-hemisphere patients showing a significantly lower hit rate than controls, and right hemisphere patients showing a trend in the same direction. Across all subjects, the ability to correctly reject face-name pairs was poor when the face had been seen before, but the name was new. This problem with false positives was particularly evident in patients with right hemisphere lesions (40% correct, where 50% = chance performance). Similarly, patients with left hemisphere resections performed no better than chance. These data indicate that humans are vulnerable to significant false positive identifications when the face item in a face-name pairing has been seen before, and that this problem is exacerbated following unilateral damage to the amygdala and anterior hippocampus. 582.29/NN98 - Comparison of two rodent models of intracerebral hemorrhagic stroke *C. L. MACLELLAN1, J. PEELING2, C. EDMUNDSON2, R. BUIST2, F. COLBOURNE1; 1Dept Psychol, Univ Alberta, Edmonton, AB, CANADA, 2Department of Pharmacology, University of Manitoba, Winnipeg, MB, CANADA. The autologous whole blood and bacterial collagenase models of intracerebral hemorrhage (ICH) are routinely used to study the pathology of ICH and putative treatments. We assessed the time course of bleeding and tracked the progression of injury and behavioral deficits in the whole blood and collagenase striatal ICH models. By directly comparing models, we aim to highlight differences between the models and limitations of each in experimental ICH studies. In Experiment 1, we assessed hematoma volume using a spectrophotometric hemoglobin assay at 1, 2, and 4 h after ICH (N = 60). Blood volume significantly increased from 1 to 4 h in the collagenase model (p = 0.010), but not after blood injection (p = 0.707). In Experiment 2 (N = 20), we attempted to match the collagenase insult to the standard 100 mL whole blood model. When we used 0.2 U of bacterial collagenase the hematoma volumes at 12 h were 69.2 mL and 46.11 mL in the blood and collagenase models, respectively (p = 0.030). We did not use a larger collagenase dose in order to avoid excessive mortality. In Experiment 3, magnetic resonance imaging (MRI) was used to track the ICH over 6 weeks (N = 40). T2 weighted images were obtained at 6 h, 12 h, 2 d, 4 d, 1 wk, 2 wk, 4 wk, and 6 wk following ICH. We also assessed neurological deficits 1 - 28 days after ICH. Preliminary results indicate that blood readily dissected along white matter tracts (e.g., corpus callosum) in the whole blood model, whereas it was more restricted to the gray matter (e.g., striatum) in the collagenase model. Hemoglobin breakdown occurred faster after collagenase-induced ICH. Neurological deficits completely recovered by 2 weeks in the whole blood model, whereas recovery in the collagenase rats was more gradual. This is despite the fact that the collagenase model produced a smaller hematoma volume. Based on early MRI data and our previous work, we predict that long-term histological injury will be much smaller in the whole blood model. The fact that blood dissects along white matter tracts, especially the corpus callosum, means that functional impairments resolve quickly and/or go undetected with most behavioral tests. This and limited injury make this model problematic for long-term studies testing therapeutics. 648.11/V21 - Burst correlation structure within the locomotor cycle of the fictive cat *D. L. BOOTHE1, D. A. MCCREA2, A. H. COHEN1,3,4, T. W. TROYER1,5; 1Neuroscience and Cognitive Sciences Program, Univ. of Maryland, College Park, MD, 2Spinal Cord Research Centre, University of Manitoba, Winnipeg, MB, CANADA, 3Department of Biology, Univ. of Maryland, College Park, MD, 4Institute for Systems Research, Univ. of Maryland, College Park, MD, 5Department of Psychology, Univ. of Maryland, College Park, MD. Stochastic neural networks can produce statistically differentiable outputs that depend qualitatively on underlying network connectivity. One such network is the traditional ‘clock’ model, which has been used to simulate the output of the spinal central pattern generator (CPG) for locomotion. ‘Clock’ models use a sequential order of activation within simulated interneurons to create properly timed outputs. Previous work in our lab has shown that ‘clock’ models often contain underlying architectural symmetries observable as consistencies in the correlation structure between phases of the cycle. In particular, correlation depends on the temporal relationship of the phases rather than the identity of those phases, and all phases display similar correlations with overall cycle length. To determine the statistical properties of the biological system, population recordings from flexor, extensor and bifunctional (posterior biceps and semitendinosus, (PBSt)) motor neuron pools were obtained from nerve electroneurograms during fictive locomotion in decerebrate cats. Activity within PBSt was separated into distinct phases using a thresholding algorithm. Correlations between the lengths of the different burst phases show substantial asymmetries. Specifically, PBSt bursts occurring contemporaneously with flexor activity are positively correlated with the length of the previous phase, whereas the later biphasic PBSt burst during extensor activation is strongly negatively correlated with the preceding phase. Additionally, three phases have positive correlation with overall cycle duration, whereas correlation with a fourth phase is weak or nonexistent. Currently we are extending our analysis to include correlations between multiple motor neuron pools as well as the many qualitatively different output patterns observed in fictive locomotion. We will further explore how hierarchical models of the spinal CPG may be able to produce the asymmetric patterns of burst length correlations seen in our data. 648.12/V22 - Effects of spontaneous deletions of nerve activity during fictive locomotion in the cat on MLR-evoked postsynaptic potentials in lumbar motoneurons M. LAFRENIERE-ROULA, *D. A. MCCREA; Dept Physiology, Univ Manitoba Fac Med, Winnipeg, MB, CANADA. Fictive locomotion evoked in decerebrate cats by electrical stimulation of the mesencephalic locomotor region (MLR) is accompanied by stimulus-locked, short-latency postsynaptic potentials (PSPs) in lumbar motoneurons. MLR-PSPs are usually excitatory when recorded during the locomotor phase in which the motoneuron is depolarized and inhibitory when the motoneuron is hyperpolarized. The step-cycle related modulation of MLR-PSP amplitude and sign (Shefchyk & Jordan 1985, Degtyarenko et al. 1998) and their disappearance with spinal cord cooling (Noga et al. 2003) suggests that MLR-PSPs are evoked through spinal circuits involved in the production of locomotor output. To further examine this hypothesis, we compared MLR-PSPs evoked during periods of normal fictive locomotion with those evoked during spontaneous deletions, i.e. periods when rhythmic motoneuron depolarization and hyperpolarization is briefly interrupted (Lafreniere-Roula and McCrea, 2005). Simultaneous intracellular recordings were made from pairs of antidromically identified medial gastrocnemius (MG) motoneurons during MLR-evoked fictive locomotion in decerebrate cats. MLR-EPSPs, normally present in MG motoneurons during the extension phase of the cycle, were reduced in amplitude or became absent during deletions of extensor activity. Similarly, MLR-IPSPs, normally present in MG motoneurons during the flexion phase of the cycle, were reduced in amplitude or disappeared during deletions of flexor motoneuron activity. The modulation of MLR-PSPs was similar in pairs of motoneurons recorded simultaneously. Preliminary analysis suggests a correspondence between the magnitude of the reduction in locomotor-cycle related motoneuron depolarization and the degree to which the MLR-PSP was reduced. The mirroring of LDP and MLR-PSP changes during deletions supports the hypothesis that MLR-PSPs are related to the production of locomotion. 682.14/NN68 - Fluorescent dextran tracers and MR imaging suggest blood-brain barrier disruption in kaolin-induced hydrocephalus in rats *I. SLOBODIAN1, A. SCHELLENBERG2, M. R. DEL BIGIO1; 1Pathology, Univ Manitoba, Winnipeg, MB, CANADA, 2Pharmacology, Univ Manitoba, Winnipeg, MB, CANADA. Introduction: Hydrocephalus is a common neurological condition characterized by impairment of cerebrospinal fluid (CSF) flow with enlargement of ventricular cavities in the brain. Gadolinium diethylenetriamine penta-acetate (Gd-DTPA) is a low molecular weight (590 D) magnetic resonance (MR) image contrast agent well suited for delineating blood-brain barrier disruption. Objective: To determine blood-brain barrier permeability in non-communicating hydrocephalus. Methods: Hydrocephalus was induced by injection of kaolin into the cisterna magna of rats at 3 weeks of age. T2 weighted MR images were obtained to confirm hydrocephalus 7 days post injection. T1 weighted images were obtained prior to and after Gd-DTPA injection (0.4mmol/kg) to measure blood-brain barrier disruption. To quantify contrast enhancement, percent difference images were calculated using T1 images. Dextran tracers were then injected into the cardiac left ventricle and allowed to circulate for 60 seconds prior to sacrifice. Lysine fixable tracers (Texas Red 10,000 and Fluorescein 500,000 mol. weight) were injected (5mg/100g body weight) into four 3-week old hydrocephalic rats and four 6-week old hydrocephalic rats, as well as 3 non-hydrocephalic controls for each age. Rats were perfused with 3% paraformaldehyde, brains were removed, fixed and stained for microscopic analysis. Results: Contrast-enhanced T1 weighted images showed no significant areas of enhancement. However, quantitative analysis on calculated percent difference images showed localized areas of enhancement in hydrocephalic animals. Fluorescent microscopy analysis revealed focal dextran trace leakage in hydrocephalic rats. Conclusions: Hydrocephalic rats showed areas of enhancement on calculated percent difference T1 weighted images, corresponding with microscopic evidence of tracer diffusion. This suggests some degree of blood-brain barrier permeability in hydrocephalic animals. 717.3/B54 - Loss of electrochemical activity regulates axotomy-induced NF-kB activation in adult rat dorsal root ganglia sensory neurons *D. R. SMITH1, J. SCHAPANSKY1, R. VAN DER PLOEG1, P. F. GARDINER2, G. W. GLAZNER1; 1Pharmacology & Therapeutics, St. Boniface Hospital Research Centre, University of Manitoba, Winnipeg, MB, CANADA, 2Physiology, University of Manitoba, Winnipeg, MB, CANADA. The majority of neurons die following long-term axotomy. Adult dorsal root ganglia (DRG) sensory neurons are notable exceptions to this rule, being able to survive indefinitely after axotomy and loss of target innervation. In addition, these cells uniquely survive in defined media in the absence of exogenous neurotrophic factors. Cultured DRG sensory neurons initiate a very rapid survival program in response to axotomy, centered on activation of the transcription factor NF-kB. We have previously reported that activation of NF-kB, measured by electromobility shift assay, is critical for survival of adult rat dorsal root ganglia sensory neurons in vitro, and that NF-kB DNA binding levels increase in cultured DRG neurons within 6 hrs of dissection. In vivo, DNA binding activity of NF-kB rises in the L4 and L5 DRG within 2 hrs of sciatic nerve crush (Fernyhough, Smith, et al. J. Neuroscience, Feb. 2005). Due to the rapid nature of this reaction, we hypothesized that increased NF-kB DNA binding activity in DRG neurons might be stimulated by the loss of incoming electrochemical impulses following axotomy. To test this hypothesis rats were anesthetised and a lidocaine or cold block was applied unilaterally to an otherwise undamaged sciatic nerve. NF-kB binding activity increased significantly in the ipsilateral DRG within 2 hrs of application of lidocaine or cold blocks, similar to that seen with sciatic nerve crush. In another test rats were anesthetised and both the right and left sciatic nerves underwent crush injury. Then, on one side only an electrode was placed proximal to the crush site, which delivered a small current (0.1 V, 5 Hz) to the nerve. Electrical stimulation attenuated the crush-induced increase in NF-kB binding activity in the ipsilateral DRG, in agreement with our hypothesis. Preliminary results from in vitro electrical stimulation of DRG cultures indicate a pattern of NF-kB DNA binding similar to that seen in the in vivo study. These data indicate that the loss of the electrical impulses may be the initial mechanism for activation of NF-kB DNA binding in the DRG neurons after axotomy. 727.6/D65 - Ryanodine receptor 3 is increased in expression and function in TgCRND8 mice *C. B. SUPNET1,2, J. A. GRANT1,3, C. A. NOONAN1,2, M. B. MAYNE1,2,3; 1Institute for Nutrisciences and Health, National Research Council, Charlottetown, PE, CANADA, 2Biomedical Sciences, University of Prince Edward Island, Charlottetown, PE, CANADA, 3Pharmacology and Therapeutics, University of Manitoba, Winnipeg, MB, CANADA. Disruption of intracellular calcium ([Ca2+]i) homeostasis appears to be an early event preceding the neurodegeneration that occurs in Alzheimer’s disease (AD). It is reported that ryanodine binding is elevated in hippocampal regions of AD brain in the early stages of the disease, suggesting increased levels of ryanodine receptors (RyRs); endoplasmic reticulum (ER)-resident Ca2+ release channels that mediate Ca2+-induced Ca2+ release in muscle, neurons and other cell types. There are 3 subtypes of RyRs (1,2,3) that exist in brain and it has been shown that RyR levels are increased in expression and function in 3 different mouse models of AD, although there is no indication as to which RyR isoform(s). We have shown that RyR3 mRNA levels are elevated in cortical neurons from transgenic (Tg) CRND8 mice that carry a double mutation of the amyloid precursor protein 695 (KM670/671NL+V717F). We observed that the upregulation of RyR3 is mediated by amyloid-beta (Ab)1-42, which is elevated in Tg mice. Given such evidence, our objectives were to determine which RyR subtype is increased in Tg mice and to determine the impact of increased RyRs to [Ca2+]i homeostasis. Here we show by reverse-transcriptase PCR and western blotting that RyR3 mRNA and protein levels were significantly increased in 4-month-old TgCRND8 brain compared to non-Tg littermates, while RyR1 and 2 levels did not change. Immunofluorescence assay results confirmed that RyR3 levels were elevated in both Tg and non-Tg cortical neurons treated with Ab1-42 compared to non-Tg. We used ratiometric Ca2+ imaging techniques to show that Tg neurons had enhanced [Ca2+]i responses to RyR agonist 20mM caffeine and to 250mM glutamate compared to non-Tg. To determine the contribution of elevated RyR3-gated Ca2+ stores to the augmented [Ca2+]i response, we knocked-down RyR3 levels using siRNA techniques. After siRyR3 treatment, Tg neurons had [Ca2+]i levels similar to non-Tg neurons in response to glutamate treatment. These observations suggest that the increased levels of [Ca2+]i in Tg neurons was a direct consequence of their elevated RyR3 protein expression. This study describes a novel mechanism by which Ab1-42 augments [Ca2+]i levels by upregulating the expression and function of a specific ER Ca2+ channel. 761.1/GG2 - Diabetes does not kill cultured adult sensory neurons but does alter axon morphology and impairs axonal regeneration *E. ZHEREBITSKAYA, P. FERNYHOUGH; Pharmacology & Therapeutics, St. Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA. Type 1 diabetes in humans and in experimental animal models induces sensory neuropathy that presents as sensory loss, foot ulceration and infection. Hyperglycemia and lack of insulin contribute to diabetes-related abnormalities in polyol pathway flux, protein glycation and reduced neurotrophic support. It is believed that these factors combine to enhance oxidative stress and trigger distal nerve damage. We first tested the hypothesis that high glucose concentrations induce oxidative stress leading to sensory neuron cell death. Cultures of adult rat sensory neurons were grown in F-12 media supplemented with 10% fetal calf serum for 1 month under normal (10mM) and high (50mM) glucose concentrations and levels of cell survival, reactive oxygen species (ROS) and 4-hydroxy-2-nonenal measured. Our results showed that normal adult DRG neurons did not exhibit cell death or oxidative stress when exposed to 50mM glucose. We then tested the hypothesis that sensory neurons exposed to long term diabetes in vivo would exhibit reduced survival and axon regeneration when placed in culture. Lumbar (L1-L6) DRG neurons from age matched normal and 1-2 month streptozotocin-diabetic rats were cultured in defined F-12 media -/+ neurotrophic factors (NTFs; 1nM insulin, 1ng/ml NGF, 10ng/ml GDNF and 10ng/ml NT-3). Cell survival, axonal morphology and levels of axon regeneration were assessed at 1 and 4 days in culture. Neuronal survival was the same between normal and diabetic neurons over 4 days, however satellite cells were markedly reduced in diabetic cultures. After 1 day diabetic DRG neurons showed enhanced axonal sprouting compared with non-diabetic neurons - this was NTF-independent. At 4 days levels of NTF-induced growth were extensive and elevated compared with no NTF treatment. However, levels of NTF-induced growth were significantly reduced in diabetic cultures compared with control. Interestingly, the axonal structure of diabetic neurons exhibited abnormal growth cones, degenerating processes and swellings or accumulations of material along axons - presumably cytoskeletal in origin. In summary, high glucose concentration had little effect on normal sensory neuron survival or growth and there was no evidence of oxidative stress. However, 1-2 months of type 1 diabetes altered satellite cell survival and impacted on axon morphology and growth in DRG cultures. 761.2/GG3 - Abnormal mitochondrial respiration is normalized by insulin treatment that does not affect hyperglycemia in sensory neurons in diabetic rats Z. WANG1, D. SMITH1, G. W. GLAZNER1, N. A. CALCUTT2, *A. P. MIZISIN2, P. FERNYHOUGH1; 1Pharmacology & Therapeutics, St Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA, 2Dept Pathol (Neuropathol) 0612, Univ California Sch Med, La Jolla, CA. Diabetic sensory neuropathy is the most common form of peripheral neuropathy and involves the dying back of distal axons leading to sensory loss and, in many cases, lower limb amputation. Though the etiology remains unclear, impairment of neuronal metabolism has been proposed as a critical event in this disease. We previously reported that mitochondrial inner membrane of dorsal root ganglia (DRG) neurons of streptozotocin (STZ) diabetic rats was partially depolarized. This was prevented by treatment with insulin at a dose too low to correct hyperglycemia. We therefore tested the hypothesis that mitochondrial depolarization in diabetes was the result of impaired rates of electron transport. The activity of the mitochondrial electron transport chain was directly assessed and the impact of low dose insulin therapy, without correction of hyperglycemia, was studied. The rate of oxygen consumption was measured with a Clarke-type oxygen electrode in mitochondrial preparations derived from DRG of age-matched control, 8-12wk STZ-diabetic rats and diabetic rats treated with low dose insulin. Rates of coupled and uncoupled mitochondrial respiration were not changed after 8 wks of diabetes compared with control. However, with the progression of untreated diabetes (12wks), there was a significant increase in coupled and uncoupled rates of oxygen consumption in diabetic compared to control (p<0.05 for coupled; p<0.01 for uncoupled). This was corrected in the low dose insulin-treated diabetic rats (p<0.05 for uncoupled) despite persistent hyperglycemia and accumulation of polyols in the nerve. Tibial nerve morphometry studies showed there was a trend for a reduction in mean axonal diameter and the relative frequency of large myelinated fibers after 12wk of diabetes that was partially prevented by insulin. Therefore, mitochondrial depolarization is not caused by impaired rates of electron transport. Indeed, the results show elevated rates of mitochondrial electron transport. Low dose insulin therapy corrected the abnormality in mitochondrial respiration and the results are discussed in terms of lack of neurotrophic support and insulin-dependent normalization of impaired calcium homeostasis in sensory neurons in diabetes. 761.4/GG5 - NF-kB and CREB DNA binding activities are aberrant in sensory ganglia and peripheral nerve of streptozotocin type 1 diabetic rats *P. FERNYHOUGH, D. SMITH, J. S. SCHAPANSKY, G. W. GLAZNER; Pharmacology & Therapeutics, St Boniface Research Centre, University of Manitoba, Winnipeg, MB, CANADA. Diabetic sensory polyneuropathy is the most common form of peripheral neuropathy and involves the dying back of distal axons leading to sensory loss and, in many cases, lower limb amputation. The sural nerve, a sensory branch of the sciatic nerve, undergoes distal axonal shrinkage and small fiber loss in human diabetic patients and in animal models of type 1 diabetes. The transcription factors NF-kB and CREB mediate many events in peripheral nerve following injury, including inflammation, Wallarian degeneration, and demyelination. We report in another abstract (C. Tweed, D. Smith et al.) that NF-kB also plays a role in axonal sprouting following nerve injury. Because diabetic sensory neuropathy may share biochemical traits with injured nerves, and both NF-kB and CREB are important mediators of nerve repair following injury, we tested the hypothesis that regulation of CREB and NF-kB DNA binding activities is altered in peripheral nerve tissues in an animal model of diabetic peripheral neuropathy. Adult Sprague-Dawley rats were injected i.p. with 75mg/kg of streptozotocin, a drug which destroys pancreatic beta cells, thus rendering the animal type 1 diabetic. After 8 to 10 weeks of diabetes (confirmed by hyperglycemia) rats were euthanized and L4/L5 lumbar dorsal root ganglia (DRG), ventral and dorsal roots, and sciatic, peroneal, and sural nerves immediately removed. Electromobility shift assays were performed on whole tissue lysates for DNA binding activities of CREB and NF-kB. Both NF-kB and CREB DNA binding activities were significantly increased (125% and 172% respectively; p<0.05 for both) in sciatic nerve in diabetic vs. control rats, and there was a trend to increased NF-kB DNA binding activity in sural nerve. In contrast, NF-kB, but not CREB, DNA binding activity was significantly decreased (~ 40% of control; p<0.05) in diabetic vs. control L4/L5 DRG. No significant differences were seen in the other tissues examined for either transcription factor. These data indicate that diabetes may cause aberrant regulation of both CREB and NF-kB, two transcription factors involved in many aspects of peripheral nerve function and recovery from injury, in the peripheral nervous tissues which are most significantly impaired in diabetic peripheral neuropathy. 761.5/GG6 - Inhibition of TNFa increases early axonal regeneration and reduces NF-kB DNA binding activity following sciatic nerve injury *G. W. GLAZNER1, C. TWEED2, D. SMITH1, J. SCHAPANSKY1, P. FERNYHOUGH1; 1Pharmacology & Therapeutics, St. Boniface Res Centre, Univ. Manitoba, Winnipeg, MB, CANADA, 2School of Biological Sciences, University of Manchester, Manchester, UNITED KINGDOM. The pro-inflammatory cytokine TNFa has been implicated in many of the events involved in peripheral nerve injury and repair, including inflammation, neuropathic pain and Wallerian degeneration. Though the pro-inflammatory actions of TNFa are often mediated through activation of the transcription factor NF-kB, the regulation and function of this inflammatory axis immediately following peripheral nerve injury has not been examined. To study the role of TNFa in modulation of inflammation and axonal sprouting following peripheral nerve injury, adult rats underwent unilateral sciatic nerve crush. Electromobility shift assays (EMSA) revealed that within 6 hours post-crush NF-kB DNA binding levels increased significantly in a 1 cm section of sciatic nerve, centered on the crush site. In addition, immunofluorescent staining for NF-kB subunits verified increased nuclear localization of p50, but not p65 or c-Rel, in Schwann cells near the site of injury, with no evidence at that time of invading immune cells. To determine if NF-kB activation was due to stimulation by TNFa, rats were injected intraperitoneally (i.p.) with saline (control), 100 or 500 mg etanercept, a soluble human recombinant TNFa receptor which decreases endogenous TNFa levels. This drug is currently in use for control of arthritis in humans. Injection of etanercept significantly reduced NF-kB DNA binding activity at the injury site 3 days after sciatic nerve crush compared to animals injected with saline. In addition, immunofluorescent staining confirmed that p50 nuclear localization in Schwann cells was lower in etanercept treated vs. control nerves following nerve injury. To determine if TNFa mediated axonal sprouting, rats were subjected to unilateral nerve crush followed by injection of either saline or 500 mg etanercept, and axonal growth determined 3 days later with immunofluorescent staining for GAP43. The regeneration distance of leading axons from the site of nerve crush was significantly greater in etanercept treated animals than saline-treated controls (5.4 mm vs. 4.6 mm, respectively; p<0.05). These data indicate that inhibitors of the TNFa/NF-kb axis may be useful therapeutic interventions following nerve injury. 832.9/HH14 - Toxic effect of blood components on oligodendrocyte precursor proliferation, migration, and differentiation in vitro J. A. PACKIASAMY, E. E. FROST, *M. DEL BIGIO; Dept Pathology, University of Manitoba, Winnipeg, MB, CANADA. Introduction: The human germinal matrix is susceptible to hemorrhage after premature birth. By 23 weeks gestation (the earliest survivable birth), all neurons have been generated, but glial cells continue to be produced. We hypothesize that periventricular hemorrhage (PVH) in the immature brain prevents normal brain development by interrupting the generation of glial cells. To test the hypothesis, we determined the toxic effect of blood components on primary cultures of oligodendrocyte progenitor cells (OPCs). Methods: OPC were isolated from rat pups aged 0-1 day and cultured in DMEM containing 10% FBS for 10-14 days. OPC were isolated from the mixed glial culture by shaking at 260rpm for 16-20 hours at 37ºC. Contaminating microglia were removed by differential adhesion. The OPC were plated at density of 5x104 cells in poly-D- lysine coated glass chamber slides and cultured for 0-6 days. Cells were treated with serial dilutions of rat plasma or serum, (1:1, 1:10, 1:100, 1:1000, 1:1000, 1:10000 and 1:100000) thrombin and plasmin contained in 1:2 and 1:8 dilutions of plasma. Proliferation of OPC was assessed by bromodeoxyuridine (BrdU) incorporation. Extent of differentiation was assessed by myelin basic protein expression detected by immunohistochemistry. Cell migration was assessed using the agarose drop assay. Results: 1) Increased cell death was observed when cells were cultured with plasma and serum in high concentrations (1:1 to 1:10) while cells treated with lower concentrations had the simplified bipolar morphology of progenitors compared to controls which had the multi-processed morphology of more differentiated cells. 2) Thrombin, but not plasmin, inhibited cell proliferation and suppressed differentiation in a concentration dependent manner. 3) Plasma, serum, thrombin and plasmin significantly decreased the OPC migration at higher concentrations. Conclusion: This study demonstrates that blood components such as plasmin and thrombin adversely affect OPC proliferation, migration and differentiation. We conclude that PVH affects the development of oligodendrocytes which will ultimately affect brain development. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||