Novel structure of the human GDNF gene.

Glial cell line-derived neurotrophic factor (GDNF) is the most potent known survival factor for substantia nigra neurons, which degenerate in Parkinson's disease, for spinal motoneurons, which die in Lou Gehrig's disease (ALS), and for Purkinje neurons, the critical outflow cells of the cerebellum. Moreover, targeted deletion of the GDNF gene results in renal dysgenesis and abnormal development of the enteric nervous system. GDNF mRNA is expressed in a complex temporospatial pattern in the central nervous system and the periphery, consistent with these observations. To begin elucidating mechanisms regulating the pattern of expression of GDNF, we have cloned the human gene, and characterized the promoter. The promoter is highly GC rich, and lacks canonical CCAT-box and TATA-box motifs. It contains more than 12 binding sites for known transcription factors. These cis-elements have the potential to interact with factors regulating constitutive expression (Sp1) and developmental expression (bHLH). Moreover, the promoter contains sites for binding transcription factors which respond to environmental signals, including CREB, AP2, Zif/268, NFkB, and MRE-BP. Combinatorial actions of these transcription factors may account for the extraordinarily complex expression patterns of the GDNF gene. Importantly, we demonstrate that the hGDNF gene utilizes a promoter distinct from that identified in the rodent GDNF gene, a finding with ramifications for Parkinson's disease and ALS research.

The identification of a novel cDNA preferentially expressed in the olfactory-limbic system of the adult rat.

To analyze cell-specific brain gene expression, we have developed a PCR-based subtractive hybridization cloning method utilizing trace starting material, allowing isolation of novel genes expressed under specific conditions. Our previous studies indicated that local substantia nigra (SN) type 1 astrocytes elaborate an array of trophic molecules which support the survival of SN dopaminergic neurons. Therefore, the current study focused on astrocyte gene expression utilizing a type 1 astrocyte-enriched cDNA library. We report initial characterization of a novel cDNA, designated AT1-46, that is preferentially expressed in the olfactory-limbic system of the adult rat brain. Although AT1-46 is expressed widely in the periphery, it is regulated both developmentally and in a cell-specific fashion in the brain. Structurally, AT1-46 is predicted to encode a highly alpha-helical molecule with several domains of potential coiled coil formation, and exhibits a 28% amino acid sequence identity with the intermediate filament-associated protein, trichohyalin.

Characterization of the genomic structure of the mouse APLP1 gene.

Amyloid beta protein (beta A4), the major component of the core of amyloid plaques in Alzheimer disease, is derived from the transmembrane amyloid precursor proteins (APPs). Our recent studies showed that a murine member of the evolutionarily conserved APP family, amyloid precursor-like protein 1 (APLP1), is specifically localized to the cerebral cortex postsynaptic density and may thus participate in brain synaptic function. To investigate regulatory mechanisms of APLP1 synthesis at the genomic level, we isolated and characterized genomic clones containing the mouse APLP1 gene. Sequence analysis revealed a genomic structure consisting of 17 exons and a promoter region that is devoid of apparent TATA and CCAAT boxes. The 5' region contains putative binding sites for AP-1, heat-shock protein, and Sp1, suggesting that multiple elements are potentially involved in regulating transcription of the APLP1 gene.

Developmental regulation of neurotrophin-3 and trk C splice variants in optic nerve glia in vivo.

Extensive evidence indicates that the survival and development of neurons is dependent on neurotrophins. However, the factors potentially required for glial development and function, and the sites of synthesis, are not well defined. To investigate the potential role of neurotrophins in glial development in vivo, we studied the trk family of receptors and their cognate neurotrophins in the postnatal rat optic nerve using reverse transcription-polymerase chain reaction. Our results indicate that trk A, B and C messenger RNAs are expressed throughout development, and in adulthood. Both trk B and trk C expression decreased during development. However, trk C expression decreased most markedly, reaching barely detectable levels by day 90. These findings suggest that neurotrophins can affect both immature and mature glial function and that their actions may be regulated through the modulation of putative receptors. To determine whether alternatively spliced forms of trk C potentially mediate neurotrophin-3 actions, we assessed expression of the different trk C isoforms. We employed reverse transcription-polymerase chain reaction using primers that selectively amplify the extracellular or intracellular domains. Optic nerve expressed both the full-length receptor and one form containing an insertion in the tyrosine kinase domain. In addition, the expression of the insert splice variant was developmentally regulated. Our observations suggest that, in glia, actions of neurotrophin-3 are probably mediated through the full-length receptor and that selected alternatively spliced forms may also be involved in trk C receptor function. To determine whether glia at different stages of differentiation elaborate neurotrophins, we analysed expression of nerve growth factor, brain-derived neurotrophic factor, neurotrophins-3 and -4/5 in the optic nerve during development. Messenger RNAs for all the neurotrophins were detected at all postnatal ages, suggesting that progenitor cells, immature and mature glia are potential sources of neurotrophins. However, neurotrophin expression was not developmentally regulated. The invariant neurotrophin messenger RNA levels, and the changing expression of trk B and trk C during ontogeny, suggest that trophic regulation of glial development is primarily governed through modulation of receptor expression.

Multiple astrocyte transcripts encode nigral trophic factors in rat and human.

The recent discovery of glial cell line-derived neurotrophic factor (GDNF) identified a novel trophin that selectively increases survival of substantia nigra dopaminergic neurons, which degenerate in Parkinson's disease. Our previous studies indicated that GDNF RNA can be amplified from cultured rat nigral type 1 astrocytes and from rat striatum in vivo, implying local as well as target trophic support. The current study establishes the regional pattern of GDNF RNA expression in adult human brain. Reverse transcription-polymerase chain reaction (RT-PCR) analysis revealed the highest expression of GDNF mRNA in the human caudate, with low levels in the putamen and no detectable message in the nigra, suggesting that GDNF is a target-derived factor in humans. We also report the isolation of two additional GDNF-related cDNAs, termed astrocyte-derived trophic factors (ATF), which apparently result from differential RNA processing. Sequence analysis of rat ATF-1 revealed a 78-bp deletion corresponding to a loss of 26 amino acids within the prepro region of the predicted GDNF protein. The RNA processing events responsible for ATF-1 formation in rat brain are conserved in humans; we report the isolation of a full-length human ATF-1 homologue. We identified a second alternative transcript, human ATF-2; the transcript encodes a protein which differs in its first 18 amino acids from the predicted mature GDNF and ATF-1 proteins and shares the terminal 115 residues with the other two forms. To begin assessing the biologic significance of multiple transcript expression we characterized the actions of COS-expressed GDNF and ATF-1 cDNAs.(ABSTRACT TRUNCATED AT 250 WORDS)

Embryonic sensory development: local expression of neurotrophin-3 and target expression of nerve growth factor.

Development and maintenance of peripheral sensory and sympathetic neurons are regulated by target-derived neurotrophins, including nerve growth factor (NGF). To determine whether trophins are potentially critical prior to and during target innervation, for neuronal survival or axon guidance, in situ hybridization was performed in the rat embryo. We examined the expression of genes encoding NGF, neurotrophin-3 (NT-3), and their putative high-affinity receptors, trk A and trk C, respectively. Trks A and C were detected in dorsal root sensory ganglia (DRG) on embryonic day 12.5 (E12.5), implying early responsiveness to NGF and NT-3. NGF mRNA was expressed in the central spinal cord target and by the peripheral somite, at this early time, which thereby may function as a transient "guidepost" target for sensory fibers. Somitic expression was transient and was undetectable by E17.5. NT-3 was expressed in the DRG itself from E13.5 to 17.5, suggesting local transient actions on sensory neurons. NT-3 was also expressed in the ventral spinal cord at low levels on E13.5. We examined the trigeminal ganglion to determine whether cranial sensory neurons are similarly regulated. Trk A was detected in the trigeminal ganglion, while NGF was expressed in the central myelencephalon target, paralleling observations in the DRG and spinal cord. However, NT-3 and trk C were undetectable, in contrast to DRG, suggesting that the environment or different neural crest lineages govern expression of different trophins and trks. Apparently, multiple trophins regulate sensory neuron development through local as well as transient target mechanisms prior to innervation of definitive targets.

Regional and cell-specific expression of GDNF in rat brain.

The survival of ventral mesencephalic substantia nigra (SN) dopamine neurons, which degenerate in Parkinson's disease, is enhanced by glial cells in vitro. The recent isolation of glial cell line-derived growth factor (GDNF), a molecule with apparently selective effects on dopamine (DA) neurons in vitro, raises the question of whether this factor is found in normal brain cells. In this study, the polymerase chain reaction (PCR) was employed to determine the regional distribution and cellular localization of GDNF in the rat central nervous system. GDNF was expressed by SN and basal forebrain Type 1 (T1) astrocytes, with trace transcript levels present in cortical T1 astrocytes. Neuronal cultures of embryonic SN also expressed GDNF. Regionally, postnatal striatum contained the highest GDNF mRNA levels in vivo under the PCR conditions employed. Our data suggest a role for GDNF in both local and target-derived support of DA neurons, as well as potential involvement in the support of other neuronal populations in vivo.

Alternative splice forms of the murine T-cell receptor beta transcript in a cytotoxic T-cell line.

The processing of the primary transcript of the murine T cell receptor c beta region can result in two distinct forms of protein. The optional exon, C beta O, is appended at the 5' end of the constant region and to date has been observed at low frequencies in cDNA clones from heterogeneous populations of T lymphocyte. Because individual cell lines were not analyzed in those studies, it was not known whether small numbers of T lymphocytes use the C beta O exon exclusively or instead most T-lymphocytes use the C beta O exon but at low levels. We have determined that our CTL clone, B6.cl 4, produces both C beta O+ and C beta O- functional RNA transcripts. To determine the C beta O usage in this CTL clone, we coupled cDNA synthesis with the polymerase chain reaction using oligonucleotides corresponding to sequences at the V beta 14 region and sequences corresponding to C beta O or C beta 1. The data indicate that B6.cl 4 CTL clone is able to use either splice site but uses the C beta O exon at a frequency of approximately 2-3%.