Fragile X syndrome is caused by a loss of expression of the fragile X mental retardation protein (FMRP). FMRP is a selective RNA-binding protein which forms a messenger ribonucleoprotein (mRNP) complex that associates with polyribosomes. Recently, mRNA ligands associated with FMRP have been identified. However, the mechanism by which FMRP regulates the translation of its mRNA ligands remains unclear. MicroRNAs are small noncoding RNAs involved in translational control. Here we show that in vivo mammalian FMRP interacts with microRNAs and the components of the microRNA pathways including Dicer and the mammalian ortholog of Argonaute 1 (AGO1). Using two different Drosophila melanogaster models, we show that AGO1 is critical for FMRP function in neural development and synaptogenesis. Our results suggest that FMRP may regulate neuronal translation via microRNAs and links microRNAs with human disease.

Fragile X syndrome carriers have FMR1 alleles, called premutations, with an intermediate number of 5' untranslated CGG repeats between patients (>200 repeats) and normal individuals (<60 repeats). A novel neurodegenerative disease has recently been appreciated in some premutation carriers. As no neurodegeneration is seen in fragile X patients, who do not express FMR1, we hypothesize that lengthened rCGG repeats of the premutation transcript may lead to neurodegeneration. Here, using Drosophila melanogaster, we show that 90 rCGG repeats alone are sufficient to cause neurodegeneration. This phenotype is neuron specific and rCGG repeat dosage sensitive. Although devoid of mutant protein, this neurodegeneration exhibits neuronal inclusion bodies that are Hsp70 and ubiquitin positive. Overexpression of Hsp70 could suppress the neurodegeneration. These results demonstrate that neurodegenerative phenotype associated with fragile X premutation is indeed caused by the lengthened rCGG repeats and provide the first in vivo experimental demonstration of RNA-mediated neurodegeneration.

The secreted protein products of the hedgehog (hh) gene family are associated with local and long-range signalling activities that are responsible for developmental patterning in multiple systems, including Drosophila embryonic and larval tissues and vertebrate neural tube, limbs and somites. In a process that is critical for full biological activity, the hedgehog protein (Hh) undergoes autoproteolysis to generate two biochemically distinct products, an 18K amino-terminal fragment, N, and a 25K carboxy-terminal fragment, C (ref. 16); mutations that block autoproteolysis impair Hh function. We have identified the site of autoproteolytic cleavage and find that it is broadly conserved throughout the hedgehog family. Knowing the site of cleavage, we were able to test the function of the N and C cleavage products in Drosophila assays. We show here that the N product is the active species in both local and long-range signalling. Consistent with this, all twelve mapped hedgehog mutations either affected the structure of the N product directly or otherwise blocked the release of N from the Hh precursor as a result of deletion or alteration of sequences in the C domain.

Cell-type specification in the Drosophila compound eye begins at the morphogenetic furrow. The furrow sweeps across the developing eye epithelium and is coincident with four classes of cellular events: coordinated changes in cell shape, changes in gene expression, synchronization of the cell cycle, and the specification of a regular array of ommatidial founder cells. The molecular mechanisms that induce these events in the developing eye have hitherto been unknown. We identify here a gene specifically required for furrow progression, hedgehog (hh). We show that hh expression posterior to the morphogenetic furrow is continuously required for its progression. We propose that forward diffusion of hh protein induces anterior cells to enter the furrow.