BIOLOGICAL OVERVIEW

Mammalian Fragile X mental retardation gene (FMR1) encodes an RNA binding protein that acts as a negative translational regulator. A Drosophila fragile X syndrome model has been developed using loss-of-function mutants and overexpression of the FMR1 homolog (Fmr1). Drosophila Fmr1 nulls display enlarged synaptic terminals, whereas neuronal overexpression results in fewer and larger synaptic boutons. Synaptic structural defects are accompanied by altered neurotransmission, with synapse type-specific regulation in central and peripheral synapses. These phenotypes mimic those observed in mutants of microtubule-associated Futsch. Immunoprecipitation of Drosophila Fmr1 shows association with Futsch mRNA, and Western analyses demonstrate that Fmr1 inversely regulates Futsch expression. Fmr1;futsch double mutants restore normal synaptic structure and function. It is proposed that Fmr1 acts as a translational repressor of Futsch to regulate microtubule-dependent synaptic growth and function (Zhang, 2001).

Fragile X syndrome (FraX) is the most common inherited disease causing mental retardation. The defect is a trinucleotide CGG expansion in the regulatory region of fragile X mental retardation 1 (FMR1), causing transcriptional silencing and loss of the gene product, FMRP (Verkerk, 1991; Verheij, 1993). FMRP is widely expressed in fetal and adult tissues, with pronounced expression in brain and testis where major symptoms are manifested (Devys, 1993). FMRP is predominantly in the cytoplasm with occasional nuclear staining (Devys, 1993; Verheij, 1993). FMRP contains nuclear localization (NLS) and export (NES) signals (Eberhart, 1996), suggesting that it functions as a nucleo-cytoplasmic shuttle protein. FMRP contains three RNA binding domains: two K homology (KH) domains and one RGG box (Ashley, 1993a; Siomi, 1993). FMRP binds ~4% of human fetal brain mRNA in vitro, but the targets are largely unknown, except its own mRNA and myelin basic protein mRNA (Ashley, 1993a; Brown, 1998). FMRP associates with polyribosomes (Khandjian, 1996; Tamanini, 1996; Feng, 1997a) and functions as a negative translational regulator (Laggerbauer, 2001; Li, 2001; Schaeffer, 2001) (Zhang, 2001).

FraX neurological pathogenesis has attracted intensive analysis. Cerebral cortical autopsies from FraX patients show abnormal neuronal dendritic spine morphology, postulated to be associated with synaptic immaturity (Hinton, 1991, Irwin, 2001). In FMR1 knockout mice, longer and denser dendritic spines are observed, consistent with the human phenotype (Comery, 1997; Nimchinsky, 2001). FMRP is observed at synapses in the developing rat brain (Weiler, 1997) and is present in mouse brain synaptosomes (Feng, 1997b; Tamanini, 1997). Furthermore, FMRP mRNA associates with translational complexes in synaptic subcellular fractions, and the expression of FMRP is increased within minutes of glutamate receptor stimulation, suggesting that FMRP acts as a synaptic activity-dependent translational regulator (Weiler, 1997; Jin, 2000). These different lines of evidence suggest that the underlying mechanism of mental retardation in FraX patients is the result of defective synapse development or function (Zhang, 2001).

A Drosophila FraX model has been generated to specifically address the hypothesis that FMRP regulates synaptic development and function. Wan (2000) identified the Drosophila homolog of FMR1. Drosophila Fmr1 (Drosophila fragile X related) has been mutated and its roles in synaptic development and function has been assayed in two model systems in Drosophila: the eye and the neuromuscular junction (NMJ). The level of Fmr1 protein has been shown to regulate both synaptic structure and function. The Fmr1 synaptic phenotypes mimic defects observed in mutants with altered levels of Futsch, a microtubule-associated protein with homology to mammalian MAP1B. Fmr1 associates with Futsch mRNA and negatively regulates Futsch expression. Most importantly, a Fmr1;futsch double mutant restores the Fmr1 synaptic structural and functional defects in the eye and NMJ. These results suggest that Fmr1 is acting as a translational repressor of Futsch to regulate the synaptic microtubule cytoskeleton and that Futsch misregulation is sufficient to explain both synaptic structure and function defects characterizing the Drosophila FraX model (Zhang, 2001).

It is now clear that Drosophila contains a single, functionally conserved member of the FMR1 family, Fmr1 (Wan, 2000 and Zhang, 2001), compared to the three related genes present in mammals. The molecular characteristics, cellular and subcellular expression pattern, and functions of Drosophila Fmr1 and mammalian FMRP show extensive parallels. Most importantly, Fmr1 mutant phenotypes are consistent with the synaptic defects associated with human FraX patients and FMR1 knockout mice. These observations suggest Drosophila is an attractive genetic system to model FraX (Zhang, 2001).

At a gross level, lack of Drosophila Fmr1 and mammalian FMRP have similar consequences. In both cases the gene is not essential; null mutants are adult viable with a normal developmental time course. Behaviorally, both Drosophila and mammalian mutants show locomotory deficits. Although a direct comparison cannot be drawn between flight defects in the Fmr1 mutants and movement abnormalities in FraX patients, it is interesting to note that both display impaired motor control. FraX patients have visuospatial defects and Drosophila Fmr1 mutants show decreased photoreceptor function in the retina. All of these common defects can be readily explained by impaired synaptic development or function (Zhang, 2001).

Recent studies indicate that mammalian FMRP is present at synapses and regulates synaptic structure (Comery, 1997; Feng, 1997b; Tamanini, 1997; Weiler, 1997; Nimchinsky, 2001). Similarly, Drosophila Fmr1 is highly expressed in both pre- and post-synaptic neurons, as well as in postsynaptic muscles, and regulates synaptic structure. Overgrowth of dendritic spines, sites of synaptic input, is a diagnostic characteristic in FraX patients (Hinton, 1991; Irwin, 2001) and also is the primary phenotype of FMR1 knockout mice (Comery 1997; Nimchinsky, 2001), suggesting a common synaptic basis of the disease. Similarly in the Fmr1 null, NMJ synaptic terminals are overgrown, containing more arboreal branches and more synaptic boutons. It is not presently known whether human patients and FMR1 knockout mice show similar NMJ defects. In addition, it has been found that Fmr1 overexpression has the opposite and complementary consequence of inhibiting synaptic growth and arborization. Thus, synaptic growth, branching, and bouton differentiation are negatively regulated proportional to Fmr1 levels (Zhang, 2001).

Fmr1 is also a key regulator of synaptic function. Different functional/chemical classes of synapses respond differently to Fmr1 misregulation. In the eye, histaminergic photoreceptor neurotransmission is equally impaired by either loss or overexpression of Fmr1, demonstrating that a precise level of the protein is required to maintain synaptic function. At the peripheral glutamateric NMJ, in contrast, neurotransmission is strikingly enhanced by either loss or overexpression of Fmr1. The role of Fmr1 is primarily presynaptic, mediating synaptic vesicle fusion probability. It is not currently known why the polarity of Fmr1 regulation differs between these central and peripheral synapses (Zhang, 2001).

Taken together, these results strongly support a Fmr1/FMRP synaptic function: Fmr1 and FMRP are similarly expressed in pre/postsynaptic cells, play a conserved role in dendritic spine/synapse structural regulation, and Fmr1, at least, is required for differential regulation of synaptic neurotransmission. It is suggested that the FMRP family plays a conserved role in synaptic development and function, which likely underlies the behavioral and developmental symptoms of FraX patients (Zhang, 2001).

The expression of FMRP is increased locally following glutamate receptor stimulation, suggesting that FMRP acts as a synaptic activity-dependent translational regulator (Weiler, 1997; Jin, 2000). Recent evidence has shown that FMRP is a negative translational regulator (Laggerbauer, 2001; Li, 2001; Schaeffer, 2001). Given these studies, it is hypothesized that Fmr1 may act as a translational repressor mediating the coupled regulation of synaptic structure and function. Several lines of evidence suggested that Futsch, a microtubule-associated MAP1B homolog, may be a target for Fmr1 translational regulation in the Drosophila nervous system. Futsch is required for dendritic and axonal development, as well as for synaptic growth. Moreover, futsch mutants alter Drosophila NMJ architecture in a fashion similar to Fmr1 NOE (neuronal overexpression of Fmr1) animals. Misregulation of the microtubule-based synaptic cytoskeleton appears to be a likely candidate for the coupled structural and functional defects observed in Fmr1 mutants (Zhang, 2001).

Evidence is presented that Fmr1 negatively regulates Futsch expression. (1) Fmr1 associates with Futsch mRNA. This interaction is specific, since Fmr1 fails to bind other targets such as alpha-tubulin mRNAs and the interaction is missing in Fmr1 null mutants. (2) In Fmr1 null mutants, Futsch protein level in the nervous system is increased and Fmr1 neuronal overexpression causes Futsch expression to be reduced. These results show that the level of Futsch in the nervous system is inversely regulated by the level of Fmr1. Taken together, the biochemical association between Fmr1 protein and Futsch mRNA and the inverse regulation of Futsch expression by Fmr1 strongly support a hypothesis that Fmr1 acts as a negative regulator of Futsch translation (Zhang, 2001).

Futsch appears to be the major target for Fmr1 in the regulation of synaptic structure and function. Structurally, futsch hypomorphs display fewer and enlarged NMJ synaptic boutons with dispersed, punctate anti-Futsch immunoreactivity, a phenotype indistinguishable from that caused by overexpression of Fmr in dfxr^NOE. However, futsch^NOE causes synaptic overgrowth, a phenotype similar to Fmr1 null mutants. Functionally, all four genotypes (loss and overexpression of either Fmr1 or Futsch) enhance neurotransmission at the larval NMJ, and all four genotypes impair neurotransmission in the adult eye. Thus, the expression alterations of Futsch are sufficient to explain the synaptic phenotypes of Fmr1 mutants (Zhang, 2001).

The most conclusive experimental result is the suppression of Fmr1 synaptic phenotypes by the Fmr1;futsch double mutants. The double mutant develops normal synaptic architecture, including the normal number of arboreal branches and synaptic boutons. Strikingly, the double mutant reduces NMJ transmission to suppress the peripheral synaptic phenotype, while at the same time it increases photoreceptor transmission to suppress the central synaptic phenotype. Based on these results, it is proposed that the major function of Fmr1 is the negative regulation of Futsch in the nervous system, which in turn regulates microtubule-dependent synaptic structure and function. Of course, it remains probable that Fmr1 is translationally regulating multiple proteins. However, the Futsch misregulation is sufficient to explain the synaptic phenotypes in Fmr1 mutants and, by extrapolation, possibly the mental retardation of FraX patients (Zhang, 2001).

GENE STRUCTURE

Extensive alternative splicing produces different isoforms of human FMRP (Ashley, 1993b). Comparison of multiple Drosophila Fmr1 EST sequences and genomic sequence demonstrates that alternative splicing occurs across the gene. 5' UTR alternative splicing and 3' alternative polyadenylation were found. The 1 kb difference of the two major bands detected in Northern blots (Wan, 2000) is consistent with the differential 3' polyadenylation. Alternative splicing in the coding region resulting in three extra amino acids in the second KH2 domain was also noted. To date, the functional significance of this alternative splicing is unclear (Zhang, 2001).

cDNA clone length - 3250

Bases in 5' UTR - 422

Exons - 12

Bases in 3' UTR - 782

PROTEIN STRUCTURE

Amino Acids - 681

Structural Domains

Drosophila Fmr1 encodes a protein of 681 amino acids. The amino acid sequence alignment of Fmr1 to hFMR1, hFXR1, and hFXR2 reveals a significant degree of conservation among all four proteins. This is particularly noticeable in the regions corresponding to previously delineated functional domains and their relative orientations within the primary structure of the proteins. Notably, Fmr1 contains all of the key structural elements of FMR1/FXR proteins that are involved in RNA binding. The two KH domains are nearly 75% identical and 85% similar between Fmr1 and hFMR1. Highly conserved isoleucine residues implicated in KH domain function are present in the Fmr1 and the FMR1/FXR KH domains. The RGG box, the other type of RNA-binding motif found in hFMR1, is also found in Fmr1. A 40-amino-acid region that mediates protein-protein interactions among FMR1/FXR proteins is also highly conserved, showing about 50% identity with Fmr1. Finally, a leucine-rich region, which has been shown to be involved in the binding of FMR1/FXR proteins to the 60S ribosomal subunit, is also highly conserved in Fmr1. This region may serve to determine the subcellular localization of FMR1, because isoforms lacking this domain are localized to the nucleus rather than the cytoplasm. Examination of this leucine-rich sequence in hFMR1 revealed a potential HIV Rev-protein kinase inhibitor (PKI)-type NES that consists of four critically spaced, large hydrophobic amino acids, including leucine, isoleucine, methionine, and valine. FXR1 and FXR2 also contain this putative signal. In Fmr1, this leucine-rich region exhibits 70% overall sequence identity and 80% similarity to the human homologs. However, one of the leucine residues that is critical for NES function is changed to glutamine, suggesting that Fmr1 may lack nuclear export activity. Nonetheless, Fmr1 and its vertebrate counterparts clearly display a very high degree of conservation at the primary structure level (Wan, 2000).

In a systematic gain-of-function screen for genes involved in eye development, a P element insertion line EP(3)3517 under sev-GAL4 control produces a mild rough eye phenotype. The flanking genomic sequence of EP(3)3517 matches a group of overlapping EST (expressed sequence tag) clones with high homology to human FMR1. Whole Drosophila genome sequence search has shown no other significantly homologous genes. dFmr1 appears to be a prototype of the FMRP family that evolved to give rise to the three members of the mammalian family (FMRP, Fmr11P, and Fmr12P). Sequence comparison using CLUSTAL W shows that the Drosophila Fmr1 (AF205596) has 35% and 56% overall identity and similarity, respectively, to FMRP, 37% and 65% to Fmr11P, and 36% and 65% to Fmr12P. The N-terminal 383 amino acids (aa) of Fmr1 has a higher homology (50%/84%) than the C-terminal 298 aa to the corresponding segments of FMRP. Similar to FMRP, Fmr1 contains three RNA binding domains: two KH domains and one RGG box, a NLS and a NES (Wan, 2000 and Zhang, 2001)

date revised: 5 April 2002

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.