ABSTRACT
Using a search engine called Motifer, we searched the public database of the human genome for genes matching a consensus pattern of cysteine residues derived from members of the transforming growth factor-beta (TGF-beta) superfamily. We identified two genes (named MDF451 and MDF628) that display sequence similarity to members of the TGF-beta superfamily in the arrangement of six conserved cysteine residues. Phylogenetic analyses revealed that MDF451 and MDF628 constitute a distinct subgroup within the TGF-beta superfamily, distantly related to the GDNF subfamily of ligands. Both genes could be identified in several primate species in addition to human, including chimpanzee, gorilla, guereza, and green and gray monkey, but not in rodents or other non-primate mammals, and appear not to be present in the genomes of mouse, rat or zebrafish. RNAs for MDF451 and MDF628 were expressed at low levels within distinct regions of the human central nervous system, including adult cerebellum, adult spinal cord and fetal brain. Despite expression at the RNA level, both genes presented a transcribed upstream stop codon that would prevent translation of the TGF-beta-like reading frame. The coding potential of alternative reading frames was not immediately apparent. The two genes may represent TGF-beta-like pseudogenes that have recently appeared in evolution in a common ancestor of the primate lineage by duplication from a GDNF/TGF-beta-like ancestral gene.
Subject(s)
Evolution, Molecular , Multigene Family , Primates/genetics , Pseudogenes , Transforming Growth Factor beta/genetics , Amino Acid Sequence , Animals , Base Sequence , Humans , Molecular Sequence Data , Nervous System/metabolism , Phylogeny , Primates/metabolismABSTRACT
Ligand-induced receptor oligomerization is a widely accepted mechanism for activation of cell-surface receptors. We investigated ligand-receptor interactions in the glial cell-line derived neurotrophic factor (GDNF) receptor complex, formed by the c-Ret receptor tyrosine kinase and the glycosylphosphatidylinositol (GPI)-anchored subunit GDNF family receptor alpha-1 (GFRalpha1). As only GFRalpha1 can bind GDNF directly, receptor complex formation is thought to be initiated by GDNF binding to this receptor. Here we identify an interface in GDNF formed by exposed acidic and hydrophobic residues that is critical for binding to GFRalpha1. Unexpectedly, several GDNF mutants deficient in GFRalpha1 binding retained the ability to bind and activate c-Ret at normal levels. Although impaired in binding GFRalpha1 efficiently, these mutants still required GFRalpha1 for c-Ret activation. These findings support a role for c-Ret in ligand binding and indicate that GDNF does not initiate receptor complex formation, but rather interacts with a pre-assembled GFRalpha1- c-Ret complex.
Subject(s)
Drosophila Proteins , Nerve Growth Factors , Nerve Tissue Proteins/chemistry , Proto-Oncogene Proteins/metabolism , Receptor Protein-Tyrosine Kinases/metabolism , Animals , Binding Sites , Cell Line , Cross-Linking Reagents , Dimerization , Glial Cell Line-Derived Neurotrophic Factor , Glial Cell Line-Derived Neurotrophic Factor Receptors , Mice , Models, Molecular , Mutagenesis, Site-Directed , Nerve Tissue Proteins/genetics , Phosphorylation , Phosphotyrosine/analysis , Protein Binding/genetics , Protein Conformation , Protein Structure, Secondary , Proto-Oncogene Proteins c-ret , Structure-Activity RelationshipABSTRACT
Neurturin (NTN) and glial cell line-derived neurotrophic factor (GDNF), two members of the GDNF family of growth factors, exert very similar biological activities in different systems, including the substantia nigra. Our goal in the present work was to compare their function and define whether nonoverlapping biological activities on midbrain dopaminergic neurons exist. We first found that NTN and GDNF are differentially regulated during postnatal development. NTN mRNA progressively decreased in the ventral mesencephalon and progressively increased in the striatum, coincident with a decrease in GDNF mRNA expression. This finding suggested distinct physiological roles for each factor in the nigrostriatal system. We therefore examined their function in ventral mesencephalon cultures and found that NTN promoted survival comparable with GDNF, but only GDNF induced sprouting and hypertrophy of developing dopaminergic neurons. We subsequently examined the ability of NTN to prevent the 6-hydroxydopamine-induced degeneration of adult dopaminergic neurons in vivo. Fibroblasts genetically engineered to deliver high levels of GDNF or NTN were grafted supranigrally. NTN was found to be as potent as GDNF at preventing the death of nigral dopaminergic neurons, but only GDNF induced tyrosine hydroxylase staining, sprouting, or hypertrophy of dopaminergic neurons. In conclusion, our results show selective survival-promoting effects of NTN over wider survival, neuritogenic, and hypertrophic effects of GDNF on dopaminergic neurons in vitro and in vivo. Such differences are likely to underlie unique roles for each factor in postnatal development and may ultimately be exploited in the treatment of Parkinson's disease.
