Int. J. Devl Neuroscience 19 (2001) 701 – 709 www.elsevier.com/locate/ijdevneu
Oral sessions Brain Development and Schizophrenia Oral 1.1 The Search for Nongenetic Risk Factors for Schizophrenia: Is Low Prenatal Vitamin D a Candidate? J. McGRATH, D. EYLES, F. FERON, J. BROWN and A. MACKAY-SIM, Queensland Centre for Schizophrenia Research, Wacol; Department of Physiology and Pharmacology, Uni6ersity of Queensland and Centre for Molecular Neurobiology, School of Biomolecular and Biomedical Science, Griffith Uni6ersity, Qld 4111, Australia Schizophrenia is a group of brain disorders that affects one in 100 individuals, and contributes substantially to the burden of disease. It is generally accepted that a mix of genetic and environmental risk factors is involved in increasing the susceptibility to schizophrenia, however the aetiology of schizophrenia is still poorly understood. Several sources of evidence support the notion that the disorder is related to abnormal brain development. Recently, we have proposed that low prenatal 1,25-dihydroxyvitamin D3 may be a risk-modifying factor for schizophrenia. The hypothesis can parsimoniously explain diverse epidemiological features of schizophrenia such as the excess of winter/spring births, increased rates of schizophrenia in dark-skinned migrants to cold climates and the increased rate of schizophrenia births in urban versus rural setting. Similar to the links between maternal folate and risk of neural tube defect in the offspring, we hypothesize that low prenatal vitamin D may be a risk factor for schizophrenia. This talk will summarize the development of the hypothesis, and present preliminary data from our coordinated research plan aimed at examining this hypothesis. The Stanley Foundation supported this project.
Oral 1.2 Vitamin D3 Depletion during Development: Altered Behaviour and Brain Gene Expression in an Animal Model of Schizophrenia A. MACKAY-SIM, F. FE´ RON, J. BROWN, D. EYLES and J. McGRATH, Centre for Molecular Neurobiology, Griffith Uni6ersity, Nathan, Qld 4111; Queensland Centre for Schizophrenia Research, Wolston Park Hospital, Wacol, Qld 4076 and Department of Physiology and Pharmacology, Uni6ersity of Queensland, St. Lucia, Qld 4072, Australia PII: S 0 7 3 6 - 5 7 4 8 ( 0 1 ) 0 0 0 5 9 - 4
Epidemiological evidence indicates that low prenatal vitamin D may be a non-genetic risk factor for schizophrenia. Rat pups born to vitamin D-depleted mothers were depleted until birth, weaning or adulthood (10 weeks). Controls had normal vitamin D intake. Animals were tested for acoustic startle behaviour and open-field locomotion. Vitamin D depletion until adulthood (10 weeks) specifically disrupted pre-pulse inhibition of acoustic startle, (a behaviour which is consistently impaired in schizophrenia, an impairment which is ameliorated with neuroleptic drugs). There were no differences in startle threshold, habituation or open-field locomotion. This disruption was developmentally regulated and not evident at 5 weeks. After acoustic startle testing at 10 weeks, the rats were euthansed with CO2 and mRNA was extracted from brains and analysed using rat Affymetrix arrays. Adult brain gene expression was markedly dysregulated in animals deprived of vitamin D during development with decreased expression of genes involved in neurotransmission, cytoskeleton, cell cycle, signal transduction and endocrine function. These results show that developmental vitamin D depletion can induce long-term effects on brain gene expression and behaviour. Some of the genes affected in these animals have also been identified as the primary genes affected in gene array studies of post-mortem brain tissue from individuals with schizophrenia. These results provide clues to cellular pathways by which low vitamin D could alter brain development in schizophrenia. Supported by the Stanley Foundation.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 1.3 Vitamin D3 Regulates Neurogenesis in Rat Olfactory Epithelium D. McALPINE, C. ANACLET, F. FE´ RON and A. MACKAY-SIM, Centre for Molecular Neurobiology, Griffith Uni6ersity, Nathan, Qld 4111 and Queensland, Centre for Schizophrenia Research, Wolston Park Hospital, Wacol, Qld 4076, Australia Low prenatal vitamin D may be a risk-factor for schizophrenia, acting during brain development. Neurogenesis is preserved in the olfactory epithelium in human adult and it is altered in schizophrenia. We examined here the effects of vitamin D status on neurogenesis in rat olfactory epithelium. Olfactory epithelia were taken from adult rats depleted of vitamin D during adulthood or during development (until birth, until weaning, until adulthood) and from control adult rats with normal vitamin D status. Olfactory epithelia were also analysed from adult mice lacking the vitamin D receptor (VDR) and their wild-type littermates. Animals were euthanased with CO2. Cell proliferation was examined in sections of olfactory epithelium using an antibody to proliferating cell nuclear antigen. Cell proliferation in the olfactory epithelium was: (1) significantly increased in adults transiently deprived of vitamin D; (2) significantly decreased in adult rats chronically deprived of vitamin D and in mice lacking the VDR; (3) significantly increased in adult rats deprived only during fetal development. These results: (1) support the hypothesis that vitamin D is anti-proliferative for neurons; (2) indicate that in chronic vitamin D deprivation there are mechanisms which suppress proliferation of neuronal precursors below normal. This was reflected in a 2 – 3-fold reduction in expression of neuron-specific b-tubulin protein in mice lacking the VDR; (3) reveal that prenatal vitamin D depletion can have long-term effects extending into adulthood. It should be noted that cell proliferation in olfactory epithelium of adults with schizophrenia is significantly increased compared to controls, consistent with the hypothesis that low prenatal vitamin D contributes to the disease.
Oral 1.4 Low Maternal Vitamin D3 Disrupts Neurotrophin Expression, Mitosis and Brain Morphology D. EYLES, F. FE´ RON, J. BROWN, A. MACKAYSIM and J. McGRATH, Queensland Centre for Schizophrenia Research, Wacol, Qld 4076; Department of Physiology and Pharmacology, Uni6ersity of Queensland, St. Lucia, Qld 4072 and Centre for Molecular Neurobiology, Griffith Uni6ersity, Qld 4111, Australia Based on certain features of the epidemiology of schizophrenia, low prenatal vitamin D has recently been proposed as a candidate riskmodifying factor for schizophrenia (McGrath, J. 1999. Hypothesis: is low prenatal vitamin D a risk-modifying factor for schizophrenia? Schiz. Res. 40, 173 – 177.). To date, the biological plausibility of this candidate has rested on the data showing that: (a) vitamin D receptors are present in the developing brain; (b) vitamin D3 induces NGF and the low affinity neurotrophin receptor p75 in vitro; and (c) vitamin D3 is a potent antimitotic/prodifferentiation agent. Here we present the findings from an experiment comparing the brains of new born pups from vitamin D depleted versus control dams. Brains from vitamin D deplete neonates had proportionally thinner corticies and much larger lateral ventricles (PB 0.02, n= 12). Cellular mitosis was elevated in the dentate gyrus, basal ganglia and the
hypothalamus (PB 0.05) with a nonsignificant trend towards diminished apoptosis observed in all brain regions examined. Neonatal NGF protein and transcript levels were also decreased (PB 0.05). These data indicate that a decrease in prenatal vitamin D changes the way the brain develops. While the mechanisms underlying these changes are still unclear, it is consistent with the pro-differentiation anti-proliferation role vitamin D plays in other tissues. Maternal vitamin D levels may therefore play an important and previously unrecognised role in foetal brain development.
Oral 1.5 Impaired Apoptosis in the Vitamin D3 Depleted Embryonic Brain Correlates with a Temporal Decrease in p75 Expression R.A. BURKERT, J.J. McGRATH, D.W. EYLES, School of Biomedical Sciences, Uni6ersity of Queensland, Brisbane and Queensland Centre for Schizophrenia Research, Wolston Park Hospital, Wacol, Australia Vitamin D is a powerful differentiating signal throughout the body, but its effects in the brain are poorly understood. There is evidence to suggest that vitamin D regulates p75 (the low affinity neurotrophin receptor) in vitro and that this receptor may be involved in apoptotic processes, particularly later in development (Barrett and Bartlett, 1994 PNAS 91, 6501 – 6505). We investigated general measures of mitosis and apoptosis, p75 and vitamin D receptor (VDR) expression in the brains of embryos maternally depleted of vitamin D. Female Sprague – Dawley rats were deprived of vitamin D and entered into a timed breeding program. At embryonic days 13, 15, 17, 19, 21 and 23, dams were euthanised by CO2 inhalation, embryos decapitated and brains removed. P75 immunoreactivity was drastically reduced in the brains of the depleted group from E19 onwards (P B0.001, Fishers Exact). Transcript levels of p75 were also decreased in depleted animals. Consistent with the timing of these changes was a decrease in the rate of apoptosis, most notably in the cingulate (P B0.05). At this same embryonic period we also show that VDR protein levels rapidly increase in the brain and a general marker for mitosis, proliferating cell nuclear antigen, decreases. These data suggest that not only is vitamin D involved in promoting apoptosis in the brain, but that diminished p75 expression may be a possible mechanism for this process and that the period from E19 onwards is crucial for the actions of vitamin D in brain development.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 1.6 Prenatal Hypoxia Produces Long-term Brain Changes: Implications for Schizophrenia A.E. REHN 1, S. DIENI 1, D. COPOLOV 2 and S. REES 1, 1Department Anatomy and Cell Biology, Uni6ersity of Melbourne, Park6ille, Vic. 3010 and 2Mental Health Research Institute of Victoria, Park6ille, Vic. 3056 Evidence suggests that abnormalities in brain development might contribute to an individual’s susceptibility for schizophrenia. Enlargement of the lateral ventricles is consistently reported in schizophrenic patients, whilst hippocampal volume changes and alterations in the distribution of the cerebral vasculature have also been observed. Previously we demonstrated that chronic placental insufficiency (CPI) during the second half of gestation in the guinea pig results in alterations to brain structure near term (Mallard et al., 1999. Schiz. Res. 40/1, 11 – 21; Rees et al., 1997. Br. Res. Dev. Br. Res. 103/2, 103–118). Here we aimed to determine whether deficits persist postnatally. At mid-gestation, pregnant sows were anaesthetized (ketamine, 40 mg/kg and xylazine, 6 mg/kg. i.m.) and CPI was induced via unilateral uterine artery ligation. At 8 weeks of age offspring were euthanised (Na pentobarbitone, 130 mg/kg), perfused with fixative and structural analyses performed. In prenatally-compromised (PC) animals enlargement of the lateral ventricles (PB 0.05) persisted at 8 weeks of age and the percentage of parenchyma occupied by blood vessels in the frontal cortex was greater (P B0.05) than in control animals. However, the total volume of the hippocampus and of specific neuropil layers was not different from control animals. Persistent enlargement of the lateral ventricles and alterations to the vasculature indicate that some aspects of brain development relevant to schizophrenia are vulnerable to adverse intrauterine events and provide support for the neurodevelopmental hypothesis of schizophrenia.
Neuronal Survival Factors and Signalling Pathways Oral 2.1 Neurotrophin-3 is Essential for the Development and Maintenance of Subsets of Enteric Neurons A. CHALAZONITIS, D.T. PHAM, T.P. ROTHMAN, P.S. DISTEFANO, M. BOTHWELL, J. BLAIRFLYNN, L. TESSAROLLO and M.D. GERSHON, Department of Anatomy and Cell Biology, Columbia Uni6ersity, New York, NY; Milennium Pharm. Inc., Cambridge, Mass.; Department of Physiology and Biophysics, Seattle, WA and Neural De6elopment Group NCI, Frederick, MD, USA NT-3 promotes the development of enteric neurons from neural crest-derived cells (NCDC) in vitro. Because an enteric nervous system (ENS) is present in mice lacking NT-3 or TrkC we thus analyzed the physiological significance of NT-3 in ENS development. Subsets of neurons selectively expressing TrkC developed with NT-3 treatment and underwent apoptosis upon withdrawal of NT-3. Blocking antibodies to NT-3 inhibited neuronal development in cultures of NCDC mixed with non-NCDC, indicating that NT-3 is provided by
the non-crest-derived cells of the fetal gut mesenchyme. In mature rat gut, retrograde transport of 125I NT-3 was visualized following mucosal or myenteric, tertiary plexus or muscle injections. The labeling pattern suggests that the NT-3-dependent neurons located in the submucosal plexus, may be intrinsic primary afferent (IPANs) and/or secretomotor and those located in the myenteric plexus innervate other myenteric ganglia and/or the longitudinal muscles. Myenteric ganglia were hyperplasic in transgenic mice that overexpress NT-3 directed to myenteric ganglia by the promoter for D b-H. In mice lacking NT-3 or TrkC, the number of neurons were reduced in both plexuses (proximal distal; submucosal myenteric). Selective deficits of submucosal CGRP-containing neurons suggest that the late-born, CGRP-expressing IPANs neurons require NT-3 for their normal development. Animals in these studies were euthanized in accordance with the AICUC at Columbia U.; ref. Chalazonitis et al., 2001. J. Neurosci. 21, 5620 – 5636.
Oral 2.2 Transforming Growth Factor-beta (TGF-b) Regulation of Motoneuron Survival I.S. McLENNAN, W.A. DAY and K. KOISHI, Neuromuscular Research Group, Department of Anatomy and Structural Biology, Uni6ersity of Otago, Dunedin, New Zealand In most bioassays, the TGF-bs are potent survival factors for immature and mature motoneurons (Jiang, Y., Zhang, M., Koishi, K. and McLennan, I.S., 2000. J. Neurosci. Res., 62: 809 – 813). However, neutralising antibodies to TGF-bs reduce the extent of ontogenic death of motoneurons, suggesting that one of the TGF-bs is proapoptotic in vivo (Krieglstein, K., Richter, S., Farkas, L., Schuster, N., Dunker, N., Oppenheim, R.W., Unsicker, K., 2000. Nat. Neurosci., 3, 1085 – 1090). We have used a combination of techniques to investigate this paradox. Mice were anaesthetised with Ketamine and Xylazine during manipulations and were killed by cervical dislocation. Semi-quantitative RT-PCR studies revealed that TGF-b2 was the most abundant isoform in the developing limb, with the concentration of TGF-b2 mRNA falling markedly during the period of cell death. The TGF-b2 protein was predominantly located in muscle fibres. We are currently counting the numbers of motoneurons in TGF-b2 − / − mice. Two sources of TGF-b1 were detected. ELISA analysis of TGF-b1 − / − mice showed that the mother was the largest sources of TGF-b1, with a smaller source being detected in Schwann cells, by immunohistochemistry. Neonatal TGF-b1 − / − mice have normal numbers of motoneurons (Jiang, Y., 2000. PhD Thesis, University of Otago). These mice lack TGF-b1 from Schwann cells but contain maternal TGF-b1. We have further examined the role TGF-b1 by circumventing the perinatal autoimmune death of TGFb1 − / − mice by using mice with a nude background. Adult nude TGF-b1 − / − mice have normal numbers of motoneurons, which die to a typical extent after axotomy. TGF-b1 is thus not essential to either the survival or the death of motoneurons.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 2.3 The p75 Neurotrophin Receptor Enhances TrkA Signaling by Promoting the Phosphorylation of SHC and Akt G.L. BARRETT and W.R. EPA, Department of Physiology, Uni6ersity of Melbourne, Park6ille 3010, Victoria, Australia One of the most important roles of NGF is to avert inappropriate neuronal apoptosis during development of the peripheral nervous system. The optimal sensitivity of cellular responses to NGF requires the presence both of TrkA and the ‘low-affinity’ p75 neurotrophin receptor (p75NTR). The mechanisms by which p75NTR enhances NGFdependent signaling through TrkA remain far from clear. A widely held view is that p75NTR and TrkA can form a single functional receptor complex, the so-called high-affinity NGF receptor. However, the evidence for this is not clear-cut. We therefore investigated the possibility that p75NTR and TrkA interact via their signal transduction pathways. We found that p75NTR specifically enhances phosphorylation of two SHC isoforms during NGF-induced TrkA activation. Further evidence of this interaction was found downstream of SHC phosphorylation, in that serine/threonine phosphorylation of Akt was also p75NTR p75-dependent. These results do not rule out the possibility that p75NTR and TrkA can form a heterodimeric NGF receptor. They do, however, reveal the existence of additional complex interactions between the two neurotrophin receptor molecules.
Center for Neuroscience, Ben-Gurion Uni6ersity, BeerShe6a, Israel TNF-a is a cytokine produced mainly in the immune response. Recently, cytokines were demonstrated in the CNS and were shown to control neuronal functions. To study the involvement of TNFa in neuronal development and function, we have examined molecular, morphogenic and behavioral aspects in TNFa knock-out (TNFa-KO) mice compared with wild type (wt). No difference was found in newborn body weight, length, brain size, hair growth, eyelid opening and tooth eruption between both groups. Yet, higher levels of NGF were detected, by ELISA, in TNFa-KO mice as compared to wt, along the first postnatal month (P1, 7, 14, 21) but not in adult brains. Adult mice examined in Morris water maze for spatial learning, task associated with the hippocampus region. TNFa-KO mice showed significant and specific advantage in performance at the probe test (32.2 and 20.4 s, respectively, in the correct quarter). Three hours following the probe test, mice were anesthetized, sacrificed and brains rapidly removed. At this point NGF levels were greater in the homogenates of TNFa-KO mice hippocampi as compared to the wt (70 and 28 pg/mg protein, respectively) but not in homogenates of frontal cortex. We suggest that TNFa may control neuronal development and function; through regulation of NGF levels. Thus, TNFa may be important in brain regions where NGF is a key molecule for accurate circuit formation and function.
Fgf3 Acts Both Alone and in Combination with Fgf8 in Patterning Forebrain, Mid-hindbrain and Hindbrain J. WALSHE, H. MAROON and I. MASON, MRC Centre for De6elopmental Neurobiology, King’s College London, New Hunt’s House, Guy’s Campus, London SE1 1UL, UK
Regulation of the Hypothalamic–Pituitary–Adrenal Axis in the Gaz Knockout Mouse K.J. LECK 1, K.I. MATTHAEI 2, J. HOLGATE 1 and I.A. HENDRY 1, 1Di6ision of Neuroscience, 2Di6ision of Biochemistry and Molecular Biology, John Curtin School of Medical Research, ANU, Canberra, ACT 0200, Australia
We have characterised Fgf3 expression in the developing zebrafish brain and detect transcripts in anterior forebrain, mid-hindbrain boundary and hindbrain rhombomere 4 consistent with our earlier studies in chick and mouse embryos. Fgf3 and Fgf8 are co-expressed in all three locations. Antisense morpholino oligonucleotides injected into one- or two-cell stage zebrafish embryos have been used to inhibit FGF3 translation both alone and in combination with inhibition of FGF8 translation. Successful ‘gene knockdown’ by the Fgf3 morpholino (Fgf3mo) is confirmed on immunoblots using an anti-FGF3 monoclonal antibody, while the Fgf8 morpholino (Fgf8mo) phenocopies the ace zebrafish which is mutated at a splice site in the Fgf 8 gene. In the forebrain, the Fgf3mo reduces gene expression in the telencephalon and analysis of axon pathways shows that pathfinding is disrupted. Fgf3mo produces mild effects on gene expression in the mid-hindbrain region, but when applied either singly or together with Fgf8mo, more dramatic changes in gene expression are detected in the hindbrain. In addition, the otic vesicles of zebrafish injected with Fgf3mo are smaller, and are reduced or absent in zebrafish injected with both Fgf3mo and Fgf8mo. MS-222 anaesthetic is used when appropriate.
Oral 2.5 TNF alpha Regulates NGF Expression in Brains of Developing and Adult Mice H. GOLAN 1,3 and M. HULEIHEL 2, 1Department of De6elopmental Molecular Genetics; 2Department of Immunology, Faculty of Health Sciences and 3Zlotowski
Mice deficient in the alpha subunit of the heterotrimeric GTP binding protein, Gz, were similar in size to wildtype littermates at birth but became significantly smaller at 3 weeks of age. Post-weaning, the mice gradually caught up in weight over a period of 3 – 6 months (Hendry, I.A. et al., 2000. Hypertolerance to morphine in the Gaz deficient mice. Brain Research, 870, 10 – 19). This prompted us to examine the development of the hypothalamic – pituitary – adrenal axis in the Gaz knockout mice. At 3 weeks, Gaz deficient animals showed hypotrophy of the adrenal cortex, which became hypertrophic by 5 months of age. Basal levels of adrenocorticotrophin hormone (ACTH) and corticosterone were measured in plasma collected by retro-orbital bleeding in 3 weeks and 5 months old mice, under brief halothane anaesthesia. All experimental protocols have been approved by the University’s Animal Ethics Committee. We found a significantly reduced plasma ACTH levels in 3 weeks old Gaz knockout mice and a slightly elevated ACTH levels in 5-month-old mutant mice. This was accompanied by a slight decline in plasma corticosterone levels in 3 weeks old mice and a significantly increased corticosterone levels at 5 months of age. These results were consistent with the sizes of the adrenal cortex observed in the mutant animals at different ages. Our data suggest Gaz mediated signaling pathways were important in regulating the development of the hypothalamic – pituitary – adrenal axis and compensatory changes occurred when this pathway was compromised.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Neural Plasticity, Patterning and Repair Oral 3.1 Specification of a Neural Crest Lineage within the Neural Tube Y. WILSON 1, M. FORD-PERRISS 1, J.-J. PANTHIER 2 and M. MURPHY 1, 1Department of Anatomy and Cell Biology, Uni6ersity of Melbourne, Melbourne, Victoria, Australia and 2UMR 955 INRA de Genetique Moleculaire et Cellulaire, Ecole Nationale Veterinaire d’Alfort, Maisons-Alfort Cedex, France The embryonic neural crest (NC) gives rise to a large array of different cell types and one of the major questions in the field is when and how the initial decisions are made to specify NC cells to a particular fate. A prevailing view is that NC cells are initially multipotent and the decision to develop down a particular path occurs sometime after emigration of the cells. However, recent experiments suggest that NC cells may be specified to a particular lineage very early. We have investigated this question for cells of a major NC lineage, skin melanocytes. Based on our previous findings that the receptor tyrosine kinase c-kit is exclusively expressed by melanocyte progenitors within the neural crest, we have traced the fate of c-kit+ cells in the developing mouse embryo. For these studies, we employed a mutant mouse strain (WLacZ) containing the lacZ reporter gene inserted into the first exon of the c-kit gene. We find that a population of c-kit+ cells arise along the dorsal midline of the neural tube at around embryonic day 10 in these mice. These cells are strongly c-kit+ neuroepithelial cells. These cells then emigrate from the neural tube on an exclusively dorso-lateral path into the developing ectoderm. These cells acquire other markers of the melanocyte lineage and are lost in WlacZ homozygous embryos. These data indicate that c-kit is expressed on a population of neuroepithelial cells which subsequently transform into NC cells, migrate out from the tube and give rise only to melanocytes. This provides direct in vivo evidence that neural crest-derived melanocyte lineage is specified within the neural tube.
Oral 3.2 Green Lights at the Border: The Role and Fate of Neural Crest Boundary Cap Cells at Vertebrate CNS:PNS Interfaces J. COHEN 1, M. VERMEREN 1, G. MARO 2, P. CHARNAY 2 and P. TOPILKO 2, 1MRC Centre for De6elopmental Neurobiology, KCL-Guy’s Campus, London Bridge, London SE1 1UL, UK and 2Molecular ´ cole Biology of De6elopment Lab, Unite´ INSERM 368, E Normale Supe´ rieure, 46 rue d’Ulm, 75005 Paris, France Boundary cap (BC) cells are a population of neural crest-derived cells found transiently during development at interfaces between the CNS and PNS: ventrally where axons exit the neural tube, at the motor exit point (MEP) and dorsally, where primary sensory afferents enter, at the dorsal root entry zone (DREZ). By virtue of their location and early expression of neural crest lineage markers, BC cells have been implicated in the guidance of axons out of and into the neural tube. By combining immunolabelling with confocal microscopy on GFP-electroporated chick embryos we show that whilst BC cell clusters form at the DREZ before primary sensory axons enter the spinal cord, they appear after motor axons exit at the MEP.
Consistent with this difference, micro-surgical ablation of the dorsal neural tube prior to neural crest emigration, showed that motor axons are able to exit the spinal cord in the absence of neural crest cells. We have also used Krox-20 lacZ knock out mice carrying an in-frame insertion of the LacZ coding sequence to follow the fate of Krox-20 expressing neural crest cells. Krox-20 is first expressed at 10.5 dpc in the proximal dorsal and ventral roots of all cranial and spinal nerves and is initially restricted to BC cells at the DREZ and MEP until 15.5 dpc when Krox-20 is expressed in immature Schwann cells. Comparison between Krox-20 lacZ and Krox-20 Cre/R26lacZ, a Krox-20 lineage tracer line, reveals the progressive migration of BC cells along the dorsal roots into the DRG and the distal peripheral nerve. LacZ +cells persist in the DRG and differentiate into small diameter sensory neurons and satellite cells. In conclusion, we have identified a neural crest precursor population localised at the DREZ that may guide in-growing sensory afferents and are a source of pluripotential cells that contribute both neurons and glia to the DRG.
Oral 3.3 Neural Crest Plasticity and Branchial Arch Patterning P. TRAINOR, Stowers Institute for Medical Research, 1000 E. 50th Street Kansas City, MO 64110, USA Current models for craniofacial development argue the neural crest is pre-programmed and carries positional information acquired in the hindbrain to the branchial arches and peripheral nervous system. Despite the attractiveness of such a scenario, using a combination of cell transplantation, lineage tracing and in vitro culture experiments we have recently demonstrated that mouse cranial neural crest cells are in fact very plastic (Trainor and Krumlauf, 2000. Nat. Cell Biol. 2 (2), 103 – 109). Hox gene expression is independently regulated in the hindbrain, neural crest and pharyngeal arches. Consequently the pharyngeal arches do not require neural crest cells for their growth nor for their initial patterning (Gavalas et al., 2001. Development 125, 3017 – 3027). This has important evolutionary implications and we present evidence that the local environment and in particular the cranial mesoderm adjacent to the hindbrain is important for regulating both the identity and migration pathways of cranial neural crest cells. Finally our analyses of Fgf and Hox gene signalling interactions resolves the plasticity versus pre-programming paradox (Trainor et al., 2001. Science (in press).
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 3.4 Direct Autoregulation by Neural Transcription Factor Brn3a via a Highly Conserved Sensory Enhancer E.E. TURNER, M. TRIEU, S.R. ENG, A. MA and N. FEDTSOVA, Department of Psychiatry, Uni6ersity of California, San Diego and San Diego, VA Medical Center, La Jolla, CA, USA Brn3a/3.0 is a transcription factor of the POU-domain family that is expressed in the ganglia of the peripheral sensory system and in highly specific neurons in the spinal cord, hindbrain, midbrain and retina. Mice with targeted disruptions of the brn3a gene exhibit widespread loss of sensory neurons in late gestation, and neonatal death. Previously we have shown that Brn3a upstream sequences can direct beta-galactosidase expression specifically to Brn3a-expressing primary sensory neurons (‘Brn3a sensory reporter mice’), and we have used this reporter to demonstrate that mice lacking Brn3a have a marked defect in sensory axon growth which precedes neural death (Eng et al., 2001. J. Neurosci. 21, 541 –549). In recent work we have shown that beta-galactosidase expression from the murine Brn3a sensory enhancer is repressed in vivo by Brn3a in a gene dosage-dependent way. The Brn3a sensory enhancer contains Brn3a binding sequences that are highly conserved in the mouse and human brn3a gene loci, suggesting a direct mechanism for autoregulation and gene dosage compensation in Brn3a heterozygous mice. We have now verified this direct autoregulatory mechanism in vivo by generating sensory reporter mice containing point mutations of Brn3a binding sites. All animal experiments were performed with anesthesia and euthanasia according to approved institutional protocols and in accordance with the guidelines of the U.S. Public Health Service. Supported by NIH awards MH01581 and HD33442 (EET), NARSAD, Veterans’ Affairs MERIT funding and the VISN22 MIRECC.
Oral 3.5 Neural Stem Cells and Huntington’s Disease R.L.M. FAULL, W.M.C. VAN ROON-MOM, M.A. CURTIS, M. DRAGUNOW and B. CONNOR, Departments of Anatomy with Radiology and Pharmacology, Faculty of Medical and Health Sciences, Uni6ersity of Auckland, Pri6ate Bag 92019, Auckland, New Zealand The recent demonstration of endogenous stem cells in the adult mammalian brain raises the exciting possibility that these undifferentiated cells may be able to generate new neurons for cell replacement in neurodegenerative diseases such as Huntington’s disease. In the adult rat and monkey brains, neural stem cells have been identified, isolated and cultured from the subependymal region (SEP) adjacent to the caudate nucleus of the basal ganglia as well as from the hippocampus; stem cells located in both regions in animal brains have been shown to proliferate and differentiate into both neurons and glial cells and may be used as a source of replacement neurons. Although previous studies on the normal adult human brain have shown evidence of neurogenesis in the hippocampus there have been no studies showing neurogenesis or the presence of neural stem cells in the SEP of the diseased adult human brain. Using an antibody to the mitotic marker proliferating cell nuclear antigen (PCNA), we have shown an increase in thickness and the presence of numerous PCNA immunopositive cells in the SEP immediately adjacent to the caudate nucleus of the basal ganglia in Huntington’s disease com-
pared to age matched normal human brains; the increase in thickness and in the numbers of PCNA positive cells strongly correlate with the neuropathological grading and with the number of CAG repeats in the IT15 gene. These findings provide good evidence suggesting that neurogenesis may occur in the SEP adjacent to the striatum in Huntington’s disease. This suggests the possibility that neural stem cells may provide a source for cell replacement in the treatment of patients with Huntington’s disease.
Oral 3.6 Viral Mediated Delivery of Differentiation Factors to Neural Progenitor Cells S.M. HUGHES, A.I. BROOKS, F. MOUSSAVIHARAMI, P.D. STABER, S.L. SAUTER, B.L. DAVIDSON, Department of Internal Medicine, Uni6ersity of Iowa College of Medicine, Iowa City, IA Neural progenitor cells, maintained in EGF, form neurospheres that express neuroepithelial marker nestin, and differentiate into neurons and glia. Neurogenesis in the adult brain is limited, therefore strategies enhancing neuronal differentiation will improve the therapeutic use of progenitor cells. We used microarray technology to study neurosphere differentiation in response to IGF-I. RNA from cells in EGF media, and 1, 2, 4, 6 or 48 h after IGF-I induction were analysed using Affymetrix arrays. Several genes involved in the Notch/Hes pathway showed differential expression. Hes 1 and Hes 5 decreased on differentiation, while Hes 6 levels increased. Hes 6 had previously been shown to inhibit Hes 1 activity, allowing MASH 1 to activate neuroD and neurogenin expression. MASH 1, neuroD and neurogenin transiently increased with neurosphere differentiation. Gene transfer could be used to test how Hes 6 effects neural progenitor cells, however the impact of recombinant viruses on self-renewal and differentiation had not been evaluated. We tested several recombinant viruses, adenovirus (Ad), adeno-associated virus (AAV) and feline immunodeficiency virus (FIV). Ad induced glial differentiation; AAV serotypes 2, 4, and 5 did not transduce progenitor cells. FIV pseudotyped with the VSV-G envelope transduced progenitor cells and these cells retained self renewal and differentiation properties and responded to trauma in vivo, migrating into a tumour from the contralateral hemisphere. Using FIV we found that Hes 6 gene transfer to neural progenitor cells caused a dramatic increase in neuronal differentiation. These studies demonstrate the utility of microarrays coupled to FIV-mediated gene transfer. All animal procedures approved by the University of Iowa Animal Care and Use Ethics Committee.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 3.7 Adult Human Eye Melanocytes can Transdifferentiate into Neurons Y. ARSENIJEVIC, N. TAVERNEY, C. KOSTIC, L. ZOGRAFOS, D. SCHORDERET and F. MUNIER, 1 Unit of Oculogenetic, Eye Hospital, 1004 Lausanne, 2 Di6ision of Genetic Medicine, Lausanne Uni6ersity, Medical School, Switzerland Degeneration of retinal neurons leads to dramatic changes in living conditions due to vision loss. In this study we investigate the possibility of generating neurons in vitro from human eye derived melanocytes. Eyes from deceased donors constitute our tissue source. During dissection, melanocytes are either dissociated with the whole choroid tissue or scratched off the sclera to produce high melanocyte density cultures. After 1 week in growth medium containing EGF and FBS, melanocytes proliferated in clusters generating a monolayer of non-pigmented cells. After differentiation with neurotrophic factors, 0.3–0.6% of total cells presented typical neuronal morphology and expressed b-tubulin, an early neuronal marker. Throughout passages (n5 11) and donors (n =7), we observed a similar percentage of neurons. RT-PCR analysis confirmed the presence of other neuronal genes like calbindin and neurofilament. To determine whether these neurons were produced in vitro via quiescent eye progenitor recruitment or by a transdifferentiation process, we investigated the presence of precursors either in adult choroid slices or in our cultures. The transient expression of the nestin precursor marker in culture, as well as the presence of pigment in neurons, contrasting with the absence of neurons and neural precursors in in vivo choroid featured the probability of a transdifferentiation process. The cue identification leading to specific neuronal phenotypes should have potent implications for transplantation into retina and the rest of the CNS. Supported by the Swiss National Science Foundation.
Axon Guidance and Synapse Formation Oral 4.1 Migration of Neural Crest Cells into and within the Developing Lung J. TOLLET, A.W. EVERETT, M.P. SPARROW, Department of Physiology, Uni6ersity of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia Neural crest cells (NCC) are present in the foregut prior to lung bud formation at embryonic day 9.5 (E9.5, mouse) and it has been presumed that these cells will migrate with the lung buds as they grow and later differentiate into the intrinsic pulmonary neurones. To investigate this, whole fetal lungs at E10 –E14 (mice euthanised with 70% CO2/30% O2) were immuno-fluorescently processed and viewed by confocal microscopy. Antibodies to phox2b (transcription factor) and p75NTR (the low affinity neurotrophic factor) both expressed by NCC, and PGP 9.5 (pan-neuronal marker) were used. We found that NCC, positive for phox2b and p75NTR, were present only on the primitive foregut and not on the protruding lung buds. At E11 both the vagus and nerve tracts from the vagus to the trachea and bronchi contained NCC. By E12, PGP 9.5- and p75NTR-positive nerve trunks, originating from large ganglia in a nerve plexus at the hilum, extended along the lobar bronchi. These ganglia and nerve tracts contained NCC but the migration of these cells lagged behind the growth of tubules and the extension of nerve trunks, suggesting that
the NCC were being guided by the nerve pathways. To investigate which neurotrophic factor might be involved in guiding this migration, lung explants (left lobes, E12) were cultured with or without glial-derived neurotrophic factor (GDNF), or with GDNF-impregnated beads. Nerve growth was increased in response to GDNF and neurite extension was directed towards the impregnated beads. We further show that the NCC and nerves are capable of responding to GDNF since they were immuno-positive for GFRa1 (receptor to GDNF) both in vivo and in vitro. We conclude that nerves may be attracted by GDNF, possibly expressed in the vicinity of the smooth muscle-covered tubules, and that the NCC will follow these nerve pathways in their migration into and within the developing lung.
Oral 4.2 Encapsulated Olfactory Ensheathing Cells in Spinal Cord Repair M.I. CHUAH 1, D. CHOI-LUNDBERG 1, S. 1 WESTON , R.S. CHUNG 2, A. VINCENT 1, A.K. WEST 2, 1Discipline of Anatomy and Physiology and 2 Discipline of Biochemistry, Uni6ersity of Tasmania, Hobart 7001, Tasmania, Australia In recent years, olfactory ensheathing cells (ECs) injected into lesions made in the spinal cord have resulted in some structural, physiological and behavioural recovery. However, the cellular mechanisms underlying these changes are not clearly understood. Our laboratory has previously shown that cultured ECs produce several growth factors, including NGF, BDNF, GDNF and various isoforms of neuregulin. To determine whether growth factors secreted by ECs are sufficient for repair or whether direct contact between ECs and host tissues is essential, we have devised a model in which ECs are loaded into a porous polymer capsule and introduced into the injury site. Under 3% isoflurane anesthesia, the dorsal columns of the spinal cord of rats were lesioned bilaterally and a capsule containing up to 30,000 ECs, or one containing the vehicle without cells, was placed adjacent or slightly lateral to the lesion. To identify axons of the corticospinal (CS) tract, a second surgery, using the same anesthetic, was performed 20 – 22 days later. Fluoro-ruby (10%, dissolved in 0.5% Triton X-100) was injected into regions of the brain containing CS neurons, including the motor cortex and hindlimb somatosensory cortex. Two weeks after tracer injection, the rats were anesthetised with Nembutal (50 mg/kg) and perfused with 4% paraformaldehyde. Sagittal sections (60-mm thick) containing the lesion site and capsule, and cross sections located both rostral and caudal to the lesion were made. Preliminary findings suggested that rats implanted with capsules containing ECs had more fluoro-ruby-positive nerve fibres than control rats, distal to the lesion site. Some of these fibres appeared to be collateral axons sprouting from the intact ventral CS tract.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 4.3 Odorant Receptor Protein Expression in Axons and Olfactory Bulb Glomeruli Supports Their Role in Axonal Guidance V.N. DEDOV, K. STERLING-LEVIS and A.M. CUNNINGHAM, De6elopmental Neurosciences Program, School of Women’s and Children’s Health, Faculty of Medicine, Uni6ersity of New South Wales, NSW 2052, Australia The olfactory system provides a unique opportunity to study the processes of axon outgrowth and targeting mechanisms. Sensory neurons within the olfactory neuroepithelium (ON) expressing the same odorant receptor (OR) mRNA project axons terminating in one of a small number of glomeruli in the olfactory bulb, and this convergent projection is fundamental to the olfactory code. The precise mechanism by which axons navigate past an array of potential synaptic targets to select the appropriate glomerular target is not known, but knockout studies have suggested that the OR itself plays a critical role in this process (Mombaerts et al., 1996; Wang et al., 1998). To further elucidate their role, we used two polyclonal antibodies generated to recognise the family of OR proteins, one to the 2nd intracellular (IC) domain and the other to the C-terminus. We believe these antibodies distinguish between the pre-inserted IC proteins and the folded, membrane-inserted forms of ORs. Adult rats were anaesthetised with IP barbiturate and perfused with PFA for immunohistochemistry. We demonstrated that OR proteins were present in the axons of olfactory sensory neurons. Additionally, we found two patterns of immunoreactivity: the C-terminal antibody labeling cilia and filling the presynaptic compartment of the glomerulus, in a manner similar to OMP and synaptophysin. The IC domain antibody selectively labeled IC accumulations of OR in olfactory sensory neurons and their dendritic projections. In the bulb, this antibody most heavily labeled nerve fibres as they entered the glomerulus, in a distribution similar to that of GAP43. Finding the membrane-associated form of the OR proteins on axons and in the presynaptic glomerular compartment is supportive of a dual role for these receptors in chemosensation and olfactory axonal targeting.
Oral 4.4 The Direction of Neural Crest Cell Migration Influences the Projections of Early Enteric Neurons H.M. YOUNG, B.R. JONES and S.J. MCKEOWN, Department of Anatomy and Cell Biology, Uni6ersity of Melbourne, 3010, VIC In embryonic mice, vagal neural crest cells enter the developing foregut around E9.5 and then migrate rostrocaudally to colonize the entire gastrointestinal tract by E14.5. Using antibodies that recognize un-differentiated neural crest-derived cells and neuron-specific proteins, we showed previously that a sub-population of crest-derived cells, very close to the migratory wavefront, starts to differentiate into neurons early; all of these early neurons transiently express tyrosine hydroxylase (TH). In this study we examined: (i) the derivatives of the TH cells; (ii) the projections of early enteric neurons; and (iii) whether migration direction influences the projections of early enteric neurons. Time-mated adult mice were killed humanely and the embryos removed. The TH cells were demonstrated to be the progenitors of nitric oxide synthase (NOS) neurons. Immunohistochemistry, lesions and DiI tracing were used to examine the development of axon projections. The axons of the first neurons in the gut (the TH/NOS neurons) projected in the same direction (caudally), and traversed the
same pathways through the mesenchyme as the migrating, undifferentiated, vagal crest-derived cells. DiI tracing studies showed that very few rostrally (orally)-projecting neurons could be detected prior to birth. To examine if the direction of migration and axon projection direction are linked, co-culture experiments were set up in which vagal crest-derived cells migrated either rostrocaudally (as they do in vivo), or caudorostrally (which they do not normally do), to colonize explants of embryonic aneural hindgut. Axon projections were correlated with the direction of cell migration, but migration direction appears to be not the only mechanism influencing axon projection patterns.
Oral 4.5 The Motor Neurite Terminal Determines where Neuromuscular Synapses Form in the Absence of Agrin G.B. BANKS and P.G. NOAKES, Department of Physiology and Pharmacology, School of Biomedical Sciences, St. Lucia 4072, Uni6ersity of Queensland, Queensland, Australia Agrin is a proteoglycan secreted by motor neurite terminals that functions to initiate and maintain AChR clusters at the nerve terminal. This led to the theory that neurite terminals decide where neuromuscular synapses form by secreting agrin. However, initiation of AChR clustering occurs in the absence of the innervating motoneuron and in the absence of agrin. In this instance, the muscle, not the nerve, is deciding the location of neuromuscular synapses by drawing neurite terminals towards pre-existing AChR clusters. If this were true, one would expect the initial innervation patterns to be the same in agrin-deficient mice and wild-type mice. To test this we quantified the intramuscular axonal branching and synapse formation in the diaphragm at E14.5 in agrin-deficient mice and wild-type mice. Heterozygote mothers were anaesthetised with Nembutal (30 mg) and killed via cervical dislocation. In the diaphragm, the nerve trunk runs down the centre of the muscle and extends branches primarily toward the lateral side. In agrin-deficient mice however, we found significantly more branches exited the phrenic nerve trunk, branched in the periphery and extended further on the medial side. Moreover, we found that the percentage a-bungarotoxin/synaptophysin colocalisations, markers of pre- and postsynaptic differentiation, respectively, was the same in agrin-deficient mice and wild-type mice. These results show that initial innervation patterns are not the same in agrin-deficient mice and wild-type mice indicating neurite terminals, not muscle, decide the placement of neuromuscular synapses in the absence of agrin.
Abstracts/Int. J. De6l Neuroscience 19 (2001) 701–709
Oral 4.6 Rapid Developmental Regulation of Presynaptic Factors Governing Synaptic Transmission and Plasticity at Central Synapses M.C. BELLINGHAM, Department of Physiology and Pharmacology, Uni6ersity of Queensland, Brisbane, QLD, 4072 During maturation, the glutamatergic endbulb-bushy cell synapse shows rapid changes in glutamate receptor-mediated excitatory postsynaptic currents (EPSCs), due to developmental modification in preand postsynaptic factors (Bellingham, M.C., Lim, R. Walmsley, B., 1998. J. Physiol. 511, 861 –869). To investigate developmental changes in presynaptic mechanisms, whole cell patch clamp recordings of single fibre evoked EPSCs (eEPSCs) were made from bushy cells (n = 54) in slices from rats (age P4 – 21) anaesthetized with sodium pentobarbitone (20 mg/kg i.p.). Paired stimuli at 5 –140 ms intervals produced marked facilitation of the 2nd eEPSC amplitude at
P4– 7 (mean ratio 9 SEM at 10 ms=1.68 90.11, n = 17), marked depression at \P11 (0.63 90.05, n = 17), and a mixture of facilitation and depression at P8 – 10 (0.99 90.15, n =18). Depletion of the readily releasable pool of synaptic vesicles by stimulus trains (100 Hz) produced marked eEPSC amplitude depression at all ages (P4 –7, 0.03 9 0.04, n =7; P8– 10, 0.02 90.03, n =6; \P11, 0.13 9 0.08, n =13). Recovery from depletion was similar at short delays (eEPSC amplitude ratio at 230 ms after train; \ P11, 0.47 9 0.27, n= 11; BP11, 0.42 90.19, n =13) but slower in \ P11 for longer delays (ratio at 1040 ms; \P11 0.6 90.16, n = 5; BP11, 0.91 90.27, n= 6). The Ca2 + sensitivity of transmitter release and paired pulse ratio were modified during development, as halving or doubling external Ca2 + caused larger changes in eEPSC amplitude and paired pulse ratio at ages BP11. These data suggest that Ca2 + -sensitive presynaptic mechanisms regulating release and short term plasticity can be rapidly modified during synaptic development. Supported by Australian Research Council and Clive and Vera Ramaciotti Foundation.