* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download [PDF]
Survey
Document related concepts
Extracellular matrix wikipedia , lookup
Cellular differentiation wikipedia , lookup
Magnesium transporter wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
Endomembrane system wikipedia , lookup
Cytokinesis wikipedia , lookup
Hedgehog signaling pathway wikipedia , lookup
Signal transduction wikipedia , lookup
Nuclear magnetic resonance spectroscopy of proteins wikipedia , lookup
Protein phosphorylation wikipedia , lookup
Protein moonlighting wikipedia , lookup
Transcript
Developmental Cell, Vol. 8, 43–52, January, 2005, Copyright ©2005 by Elsevier Inc. DOI 10.1016/j.devcel.2004.10.020 Fragile X Protein Functions with Lgl and the PAR Complex in Flies and Mice Daniela C. Zarnescu,1 Peng Jin,2 Joerg Betschinger,4 Mika Nakamoto,2 Yan Wang,5 Thomas C. Dockendorff,6,7 Yue Feng,3 Thomas A. Jongens,6 John C. Sisson,5 Juergen A. Knoblich,4 Stephen T. Warren,2 and Kevin Moses1,* 1 Department of Cell Biology 2 Department of Human Genetics 3 Department of Pharmacology Emory University School of Medicine Atlanta, Georgia 30322 4 Institute of Molecular Pathology Dr. Bohrgasse 7 1030 Vienna Austria 5 The Section of Molecular Cell and Developmental Biology The Institute for Cellular and Molecular Biology The University of Texas at Austin Austin, Texas 78712 6 Department of Genetics University of Pennsylvania School of Medicine Philadelphia, Pennsylvania 19104 Summary Fragile X syndrome, the most common form of inherited mental retardation, is caused by loss of function for the Fragile X Mental Retardation 1 gene (FMR1). FMR1 protein (FMRP) has specific mRNA targets and is thought to be involved in their transport to subsynaptic sites as well as translation regulation. We report a saturating genetic screen of the Drosophila autosomal genome to identify functional partners of dFmr1. We recovered 19 mutations in the tumor suppressor lethal (2) giant larvae (dlgl) gene and 90 mutations at other loci. dlgl encodes a cytoskeletal protein involved in cellular polarity and cytoplasmic transport and is regulated by the PAR complex through phosphorylation. We provide direct evidence for a Fmrp/Lgl/mRNA complex, which functions in neural development in flies and is developmentally regulated in mice. Our data suggest that Lgl may regulate Fmrp/mRNA sorting, transport, and anchoring via the PAR complex. Introduction Fragile X syndrome (FraX) is the most common form of inherited mental retardation and affects about 1/4000 males, with an average IQ below 70, attention deficit and hyperactivity disorders, sleep disturbances, autism, facial dysmorphia, and macroorchidism (reviewed in O’Donnell and Warren, 2002). FraX patients have elongated neocortical dendritic spines, which represent an *Correspondence: [email protected] 7 Present address: Department of Zoology, Miami University, Oxford, Ohio 45056. immature state, associated with mental retardation (Hinton et al., 1991). A similar phenotype has been reported in FraX null mutant mice (Nimchinsky et al., 2001). The disease is a consequence of loss of function for the Fmr1 gene (O’Donnell and Warren, 2002). FMRP shuttles between the nucleus and the cytoplasm and associates with RNA via its two KH domains and an RGG box (O’Donnell and Warren, 2002). While mice and humans contain a three-member Fmr1 gene family, Drosophila has a single gene ortholog: dFmr1 (also known as dfxr; for nomenclature used here, see Experimental Procedures; Wan et al., 2000). dFmr1 mutants are viable and exhibit defects in synapse morphology and function (Lee et al., 2003; Schenck et al., 2003; Zhang et al., 2001) and in neuronal arborization and circadian rhythm (Gao, 2002). Fmrp acts at several points in RNA metabolism and has been recently shown to associate with components of the RNA interference pathway, suggesting that it may control translation via micro-RNAs (Brown et al., 2001; Carthew, 2002; Darnell et al., 2001; Jin et al., 2004). Drosophila MAP1B (futsch) mRNA is a dFmr1 target and futsch loss of function genetically suppresses a dFmr1 loss-of-function phenotype (Zhang et al., 2001). Rac1 mRNA is also a dFmr1 target, while Rac1 protein cooperates with dFmr1 in regulating dendrite arborization (Lee et al., 2003) as well as synaptic morphology via its interaction with CYFIP/ Sra (Schenck et al., 2003). In summary, the modulation of protein expression at the synapse is important for neural plasticity, and Fmrp is thought to function in this process as an agent for localized translational regulation. To identify novel genes that participate in dFmr1 function, we performed a genetic screen for dominant modifiers of retinal overexpression of dFmr1 in Drosophila. Here we report that the tumor suppressor lethal (2) giant larvae (dlgl) interacts genetically with dFmr1 in the eye neuroepithelium as well as at the glutamatergic neuromuscular synapse in a cooperative manner. dLgl is a ubiquitously expressed (Strand et al., 1994b), cytoskeleton-associated protein, which contains WD-40 domains (Betschinger et al., 2003; Plant et al., 2003). Mutations in dlgl are lethal, and phenotypic studies have demonstrated a requirement for the gene product in various aspects of development such as dorsal closure (Manfruelli et al., 1996), neuroblast delamination (Ohshiro et al., 2000; Peng et al., 2000), wing development (Arquier et al., 2001), and oogenesis (Bilder et al., 2000). dLgl contributes to the formation of molecular scaffolds by associating with itself (Jakobs et al., 1996) and other molecules including nonmuscle myosin (Strand et al., 1994a) and the PAR complex (Bilder, 2004; Strand et al., 1994b). Not only is Lgl conserved in vertebrates, but recent studies have demonstrated that its association with the PAR protein complex and its phosphorylation by atypical-PKC-zeta (aPKC-zeta) are conserved between flies and mice (Bilder, 2004). dLgl is a tumor suppressor and functions with Discs-large and Scribble in controlling cell polarity in epithelia and neuroblasts Developmental Cell 44 (Bilder, 2004; Klezovitch et al., 2004; Knust and Bossinger, 2002; Tepass et al., 2001). In yeast and vertebrate cells (MDCK), Lgl homologs regulate docking and fusion of post-Golgi vesicles with the plasma membrane through the SNARE complex (Larsson et al., 1998; Lehman et al., 1999). In summary, Lgl is believed to contribute to cellular asymmetry via its association with (1) cell junctional complexes and (2) the cytoplasmic transport machinery. Here we show biochemically that Fmrp and Lgl proteins form a functional complex in flies and mice. We also show that this complex includes specific mRNAs and that it is developmentally regulated in the mouse. Furthermore, we have data to suggest that the Fmrp/Lgl complex can be regulated by the PAR protein complex. Results dlgl Is a Genetic Suppressor of dFmr1 Gain of Function in the Developing Eye Neuroepithelium Ectopic overexpression of dFmr1 in the compound eye under the control of a sevenless gene enhancer (sev:dFmr1), results in a viable, dominant rough eye phenotype accompanied by retinal disorganization (Figures 1B and 1F; see Wan et al., 2000). To identify novel contributors to Fmrp function, we performed a chemical mutagenesis screen for autosomal dominant modifiers of the sev:dFmr1 phenotype. We screened 51,150 F1 flies and isolated 109 dominant suppressor and enhancer mutations. Of these, 63 mutations fall into 63 single-hit complementation groups and 17 are allelic to Star (a haploinsufficient locus, often identified in dominant modifier screens based on eye phenotypes). The rest (29 out of 109) fall into five complementation groups with more than one allele each. Four of these complementation groups are not yet characterized. The fifth group is notable in that we recovered 19 independent alleles, all of which are dominant suppressors of sev:dFmr1 and also recessive lethal. We mapped this group to the dlgl locus (Figures 1C and 1G; Mechler et al., 1985). As we conducted the screen, we also recovered 17 mutations in the white gene as a monitor of genetic saturation (see Experimental Procedures). These frequencies suggest that the screen was effectively saturating for loci at which a simple loss-of-function mutation interacts reliably with sev:dFmr1. dlgl is located close to a telomere, and it could be that the large number of dlgl alleles recovered in the screen might be due to subtelomeric deletion, but DNA gel blots reveal no large rearrangements at the locus in our lines (data not shown). Furthermore, we confirmed that extant dlgl mutations also suppress sev:dFmr1 (Figure 1D), as quantified by scoring the number of visible photoreceptor cells per ommatidium (see Experimental Procedures; Figures 1E–1H). Given the known roles for Fmrp in mRNA regulation and for Lgl in cell junctions and transport, the genetic interaction between dFmr1 and dlgl could suggest two possible, not mutually exclusive, models: Fmrp could act as an intermediary between specific target mRNAs and Lgl to mediate their targeted transport and/or anchoring at specific sites (such as synapses). Figure 1. dlgl Mutations Are Dominant Suppressors of sev:dFmr1 Genotypes as indicated, anterior right, dorsal up. (A–D) Adult compound eyes, scanning electron micrographs. (E–G) Retinal sections. (H) Histogram of rhabdomere numbers per ommatidium, percentage of ommatidia against number of rhabdomeres as indicated. Error bars: standard error of the mean. (I–K) Confocal sections at the same depth within larval eye discs stained for dFmr1 antigen, anterior right, arrowheads indicate morphogenetic furrow. Bracket below (J) and (K) indicates domain of transgene expression as denoted by elevated dFmr1 in sev:dFmr1 (J). dFmr1 overexpression is reduced by removing dlgl dosage (K). (L) Histogram of dFmr1 levels. Scale bars equal 50 m in (A), 10 m in (E), and 20 m in (I). We first asked how we recovered dlgl alleles in the screen: i.e., does dlgl heterozygous, loss-of-function mitigate dFmr1 overexpression? As expected, dFmr1 protein is elevated posterior to the morphogenetic furrow in sev:dFmr1 compared to wild-type (the domain of sevenless enhancer activity, compare Figures 1I to 1J). When dlgl dosage is reduced, this ectopic dFmr1 protein is reduced (Figure 1K; compare domains of transgene expression, posterior to the morphogenetic furrow). Indeed, quantification of dFmr1 staining intensity confirmed that reduction of dlgl dosage in a sev:dFmr1 background results in a 22% decrease in signal posterior Fmr Protein Acts with Lgl and the PAR Complex 45 to the furrow, where the transgene is expressed, normalized as an internal control to the anterior domain where only endogenous dFmr1 is present (compare left boxes in Figure 1J [more] to Figure 1K [less] and Figure 1L; see Experimental Procedures). These data do not suggest a mechanism, but show that the phenotypic suppression observed in the screen is simply due to a quantitative reduction in the ectopic dFmr1 protein, consistent with dlgl acting as a dominant suppressor of sev:dFmr1. dlgl Is a Dominant Enhancer of dFmr1 in Synapse Biogenesis FraX patients exhibit synaptic morphology defects, which may be interpreted as an increase in synaptic area, consistent with a reduced ability to modify synapses with learning (O’Donnell and Warren, 2002). Similarly, in flies, dFmr1 mutations affect the neuromuscular junction (NMJ) by increasing the synaptic area (Schenck et al., 2003; Zhang et al., 2001). We examined the muscle 6/7 NMJ synaptic surface (Rohrbaugh et al., 2000) and found that dlgl loss of function dominantly interacts with dFmr1. Thus, while dlgl/⫹ and dFmr1/⫹ synapses are essentially indistinguishable from wild-type, dlgl/⫹; dFmr1/⫹ larvae exhibit synaptic hyperplasia (see Figures 2A–2E). This enhancement was confirmed by counting the number of synaptic boutons, which rises more than 2-fold in trans-heterozygote individuals (mean 133 boutons, n ⫽ 16) compared to wild-type (mean 61 boutons, n ⫽ 10). This interaction is statistically significant (t test: p ⬍⬍ .001). Furthermore, these genetics are consistent with our screen. This effect is not due to an increase in muscle size, which is unchanged (see Experimental Procedures; data not shown). Thus, while the genetic screen is based on a deliberate genetic artifact (the overexpression of dFmr1 in the eye), this interaction at the NMJ supports the notion that dlgl and dFmr1 do function together in vivo. Although dFmr1 is ubiquitously expressed, dFmr1 function is required on the presynaptic side of the NMJ (Zhang et al., 2001). We asked if we can bypass the requirement for dFmr1 at the NMJ by providing additional dlgl function (see Experimental Procedures). Overexpression of a full-length dlgl cDNA, either pre- or postsynaptically, results in no significant change in the number of synaptic boutons (presynaptic: 92.7 ⫾ 2.9 [n ⫽ 10]; postsynaptic: 107.5 ⫾ 4.1 [n ⫽ 6]). This is consistent with dlgl functioning genetically either upstream or parallel to dFmr1. Is this epistatic relationship between dFmr1 and dlgl also true at the molecular level? In most places, dFmr1 is uniformly expressed at low levels (D.C.Z. and K.M., unpublished data; Wan et al., 2000). However, in the developing egg chamber, dFmr1 is upregulated in oocytes up to stage 9 (Figures 2F–2H). To avoid early developmental effects, we used a hypomorphic, temperaturesensitive allele of dlgl (dlglts3) to transiently reduce dlgl function (24 hr at restrictive temperature; De Lorenzo et al., 1999) and find that this accumulation of elevated levels of dFmr1 protein is affected (compare arrows to arrowheads in Figures 2F–2K). This is consistent with the previous finding that reduction in dlgl dosage can affect the accumulation of dFmr1 in sev:dFmr1 eye discs (Figures 1I–1L) and that dlgl acts genetically upstream or in parallel to dFmr1 function. Figure 2. dlgl Acts Genetically Upstream or in Parallel to dFmr1 (A–D) Neuromuscular junctions (NMJs), segment 3, muscle 6/7. Large (arrow) and small (arrowhead) boutons, genotypes indicated. dlgl null allele is a significant dominant enhancer of dFmr1 (compare [B], [C], and [D]). (E) Histogram: number of boutons per junction, error bars: standard error of the mean. dlgl4/⫹; dFmr13/⫹ is very highly significantly different to dlgl4/⫹ or dFmr13/⫹ alone (p ⬍⬍ .001, asterisk), while dlgl4/⫹ and dFmr13/⫹ are not statistically significant from wild-type. (F–K) Developing egg chambers stained for dFmr1 (F, H, I, K) and dLgl (G, H, J, K). dFmr1 accumulates in wild-type (F–H) but not in all dlglts3 (I–K) oocytes (compare arrows to arrowheads in [F]–[K]). Note in (J) and (K), some dLgl perdurance after 24 hr at restrictive temperature. Scale bars equal 20 m in (A) and 60 m in (F). Fmrp and Lgl Associate In Vivo to Form a Developmentally Regulated Complex In order to avoid potential artifacts associated with overexpression in tissue culture, we used immunoprecipitation experiments of endogenous proteins from living animals to show that, indeed, dLgl and dFmr1 form a protein complex in vivo (Figure 3A, asterisks). We can immunoprecipitate each protein with an antibody to the other, showing that the experiment is reciprocally replicable. However, as these experiments are not quantitative, we can draw no firm conclusions as to the stoichiometry of the complex. The presence of a dFmr1/dLgl Developmental Cell 46 but considerably stronger in 2-week-old individuals, despite the overall levels of mLgl and mFmrp levels remaining apparently constant. This is the time at which mFmrp is thought to maximally regulate synaptogenesis in the brain (Nimchinsky et al., 2001), and thus our data are consistent with a dynamic Fmrp/Lgl complex function in neural development and demonstrate a developmental component of Fmrp. In order to estimate an approximate size of the dFmr1/ dLgl complex, we subjected Drosophila extracts to sucrose fractionation (see Figure 3E and Experimental Procedures). dFmr1 and dLgl comigrate in several fractions, and based on the migration of molecular standards (data not shown), we estimate that dFmr1 associates with dLgl to form a large macromolecular complex (several hundred kilo-Daltons). Taken together, our biochemical data show that Lgl and Fmrp are present in a common complex, in vivo, in flies and mice. Figure 3. Fmrp and Lgl Proteins Coimmunoprecipitate and Cofractionate with aPKC-zeta and PAR6 Visualizing antibodies indicated right. (A) Immunoprecipitation from Drosophila heads. Input (10% of total extract), supernatant (loaded 10%), pellet, precipitating antibody indicated above. Note: anti-dFmr1 precipitates dLgl and anti-dLgl precipitates dFmr1 (asterisks). Note: for clarity, five times more dFmr1 pellet than the dLgl pellet was loaded for the dLgl blot and five times more dLgl pellet than dFmr1 pellet was loaded for the dFmr1 blot. (B) FLAG-dFmr1 precipitates dLgl from fly extracts and this is reduced by competing FLAG peptide. (C and D) Precipitating antibody indicated above; anti-mFmrp coprecipitates with mLgl in the developing mouse brain, weakly at birth (asterisk) and more strongly at 2 weeks (two asterisks). (E) Western blot of cellular fractions from a sucrose gradient show dFmr1, aPKC-zeta, and PAR6 comigrating in several fractions with and without dLgl (fractions 6–10 and 11–17, respectively). Molecular weights: dFmr1 ⵑ82 kDa, dLgl ⵑ127 kDa, aPKC-zeta ⵑ75 kDa, PAR6 ⵑ45 kDa. complex was further confirmed in vitro by anti-FLAG epitope immunoprecipitation using in vitro translated FLAG-dFmr1 protein to specifically bring down dLgl from a Drosophila extract (Figure 3B). A pull-down experiment using in vitro translated, tagged proteins failed to show that dFmr1 and dLgl bind directly (data not shown), suggesting that other (unknown) molecules may lie between Fmrp and Lgl in vivo. Next, we asked if this interaction identified in Drosophila is conserved in a mammal. Just as in the fly, we can reciprocally immunoprecipitate both endogenous mLgl and mFmrp from mouse brain extracts (Figures 3C and 3D, asterisks). Significantly, this association is developmentally regulated: it is weak in newborn mice Fmrp and Lgl Have Overlapping Expression Patterns We used immunohistochemistry to show that endogenous dFmr1 is uniformly cytoplasmic throughout the developing fly retinal neuroepithelium (Figures 4A and 4C) while dLgl is overlapping but accumulates to higher levels at the periphery of cells (Figures 4B and 4C). We have also examined other tissues (ovaries, larval neuromuscular junctions) and observed a similar result (Figure 2 and data not shown). Thus, while the endogenous dFmr1 and dLgl proteins do not have identical distributions, they are both present throughout the cytoplasm, at this level of resolution. It may be that some fraction of the cellular pool of each protein (dLgl and dFmr1) act together, while the remainder of the pool of each protein may be in some transitional state or may be contributing to some other cellular function(s). Endogenous Lgl Is Reorganized in Response to Fmrp To further test the physical interaction between Fmrp and Lgl in vivo, we used Mouse Embryonic Fibroblasts (MEFs). Endogenous mouse Fmrp (mFmrp) and mouse Lgl (mLgl) are expressed at low levels in the cytoplasm of MEFs (Plant et al., 2003). We expressed Red Fluorescent Protein fused to mFmrp (RFP-mFmrp) in MEFs (Experimental Procedures) and, as previously reported (De Diego Otero et al., 2002; Mazroui et al., 2002), we find that RFP-mFmrp assembles into perinuclear and cytoplasmic granules (arrow in Figures 4D, 4F, 4G, and 4I). Consequently, endogenous mLgl becomes concentrated in the same particles (arrow in Figures 4E, 4F, 4H, and 4I) with most RFP-mFmrp and mLgl granules clustered into mosaic-like complexes (asterisk in Figures 4G–4I). These data suggest that Lgl may act as a molecular scaffold for Fmrp granules. This shows that the endogenous dLgl protein distribution is reorganized in response to an increased level of dFmr1 protein. In converse experiments, Lgl overexpression has no detectable effect on Fmrp distribution (data not shown). These data are consistent with Fmrp being normally limiting in both systems and with Lgl protein being in Fmr Protein Acts with Lgl and the PAR Complex 47 granules in vertebrate cells but not in the fly, possibly because the vertebrate cultured cells are four to five times larger, allowing the better resolution of intracellular structures, or because we may be expressing the transgenes to different levels in the two systems. dLgl Cofractionates with dFmr1 in a Golgi Membrane-Associated Fraction Mammalian Lgl plays a role in the establishment of apical-basal polarity in epithelia by sorting basolateral-specific proteins at the level of the Golgi apparatus (Bilder, 2004). In addition, Fmrp has been found to associate with the endoplasmic reticulum (ER) in vertebrate cells (Ohashi et al., 2002). Given our observation of mFmrp/ mLgl-containing granules in the perinuclear area of MEFs and CAD cells, we investigated the distribution of dFmr1 and dLgl in a membrane fractionation experiment (see Experimental Procedures). While dLgl is widely distributed across the gradient, it cofractionates with dFmr1 in the Golgi-associated but not ER- plasma membrane-associated fractions as determined by the presence or the absence of specific markers, respectively (Figure 5A). These data are consistent with a dFmr1/ dLgl complex in association with the Golgi apparatus, although a direct link remains to be proven. Figure 4. dFmr1 and dLgl Proteins Have Overlapping Expression Proteins below, genotypes/cell type right. (A–C) Single confocal sections of developing eye, endogenous expression. (D–I) RFP-mFmrp in MEFs. (D–F) Arrow, transfected; arrowhead, untransfected cell. (G–I) Projections of seven consecutive sections (each 1 m thick) showing that the most intensely staining mFmrp granules contain mLgl (arrows), while more diffuse mFmrp granules do not (arrowheads); asterisk indicates a mosaic of both types of granules. (J–O) Mouse CNS catecholaminergic (CAD) cells. Arrows indicate colocalization of RFP-mFmrp and mLgl in granules trafficking within developing neurites. Scale bars equal 10 m in (A), 20 m in (D) and (J), and 1 m in (G) and (M). excess (see Discussion below). We also used a cell model for vertebrate neurogenesis (CAD cells, a CNS catecholaminergic cell line; Qi et al., 1997). As in MEFs, RFP-mFmrp expression results in reorganization of the endogenous mLgl into Fmrp-containing granules (Figures 4J–4O). mFmrp/mLgl colocalizing granules are visualized both in the perinuclear region and within developing neurites, suggesting a role for Lgl in Fmrp trafficking in these mouse cultured cells (arrows in Figures 4J–4O). Taken together, these data show that Fmrp can recruit Lgl to a granular complex in vivo (at least in this overexpression condition). We can detect these dLgl Associates with a Small Subset of mRNAs via dFmr1 It has been previously reported that mFmrp granules colocalize with various proteins including microtubuleassociated motors as well as RNA (De Diego Otero et al., 2002). Taken together with our data, this suggests the possibility that Lgl might associate with at least a subset of Fmrp containing RNPs and thus regulate specific mRNA targets. We used Affymetrix microarrays to identify Drosophila mRNAs that are candidate cargo in the dFmr1/dLgl protein complex. First, we selected those mRNAs that are consistently enriched by dFmr1mediated immunoprecipitated complex (compared to input mRNA). To control for specificity, we eliminated those that are equally well precipitated by an anti-dFmr1 antibody from wild-type and dFmr1 null extracts. This list of specific dFmr1-associated target mRNA consists of 83 transcripts and includes some (but not all) known targets of dFmr1 (see Figure 5A; Experimental Procedures and Supplemental Data at http://www. developmentalcell.com/cgi/content/full/8/1/43/DC1/). Next, we selected mRNAs that are enriched in Lgl-mediated immunoprecipitation (compared to input) and we identified a total of 78 Lgl-associated mRNAs. Of these, 9 were enriched in both dFmr1 and dLgl immunoprecipitations. Interestingly, none of the 9 dFmr1/dLgl shared mRNAs can be precipitated by anti-dLgl antibodies from dFmr1 null extracts, demonstrating that their association with dLgl depends genetically upon dFmr1 (see Table 1). This experimental approach is limited by absolute mRNA abundance, and thus we are less able to detect cargo mRNAs that are expressed at relatively low levels (there is a detection sensitivity threshold), so we expect that the true list of in vivo cargo mRNAs is longer. In summary, we have detected 9 mRNAs that associate specifically in a dFmr1/dLgl complex via the dFmr1 Developmental Cell 48 Taken together, our data are consistent with dLgl as a regulator of a specific set of dFmr1 functions. Figure 5. dFmr1 and dLgl Cosediment with Golgi Membranes in Density Gradients and Associate with mRNA (A) Western blots of fractions from a linear Optiprep density gradient show that the majority of membrane-associated Lgl and Fmr cosediment with the Golgi-associated protein Lava lamp (Lva). Portions of membrane-associated Lgl also cosediments with the endoplasmic reticulum (ER) marker BiP and the plasma membrane marker Toll, fractions 2–5 (data not shown). Equal volumes of each fraction and 50 g of the gradient load (P100) were analyzed. (B) Venn diagram shows the total number of genes represented on the chips (large square: 14,010), the number of genes present in input (small square: ⵑ6,250), the number of mRNAs specifically associated with dFmr1 in wild-type (83), and the number of mRNAs enriched in dLgl IP from wild-type versus input (78). Nine mRNAs are shared between dFmr1 and dLgl IP but unchanged in dLgl IP versus input from dFmr1 nulls (see Table 1). partner. Although most of these 9 genes remain to be characterized, two of them (CG9681 and CG3348) have been previously shown to cycle in synchrony with the circadian clock (McDonald and Rosbash, 2001). This is particularly interesting because dFmr1 has been demonstrated to play a role in circadian rhythms (Gao, 2002). Lgl Functions with Fmrp in the PAR Complex The PAR complex is essential for cellular polarity in many taxa. Furthermore,the PAR complex acts with dLgl in neuroblast polarity and asymmetric cell division in Drosophila (Ohno, 2001). Within the PAR complex, aPKC-zeta phosphorylates Lgl, thus rendering it inactive and changing its binding properties (Betschinger et al., 2003; Plant et al., 2003). mLgl and mPAR-6 colocalize at the Golgi in MEFs, and in response to wounding, mLgl and aPKC-zeta colocalize at the leading edge of the cells where they regulate polarized migration (Plant et al., 2003). Does the PAR complex interact genetically and/or physically with Fmrp? In sucrose gradients of larval extracts, aPKC-zeta and PAR-6 migrate in fractions that contain dFmr1 and not dLgl (Figure 3E, fractions 6–10) as well as in fractions that contain both dFmr1 and dLgl together (Figure 3E, fractions 11–17). This is consistent with a PAR complex association with dFmr1 and dLgl in the higher fractions. During oogenesis, dFmr1 accumulates in the developing oocytes, particularly at the posterior pole (Figures 2F and 6A) and colocalizes with Bazooka (Baz/PAR3; Figures 6A and 6B). aPKC-zeta is also enriched at the oocyte posterior, albeit in a manner restricted to the membrane rather than throughout the oocyte cytoplasm (data not shown; Cox et al., 2001). Furthermore, heterozygous loss-of-function mutations in both Baz (Baz4) and PAR6 (PAR6Delta226) act as dominant enhancers of sev:dFmr1 (Figures 6D–6F), while having no dominant eye defects on their own (data not shown). This suggests that PAR3/PAR6 antagonize dFmr1 function. aPKC-zeta is a core regulator of the PAR complex and controls synapses in Drosophila by regulating microtubule dynamics (Ruiz-Canada et al., 2004). A lossof-function allele for aPKC-zeta (aPKCk06403) genetically suppresses the sev:dFmr1 rough eye phenotype just as does dlgl (Figure 6D, compare with Figure 1). At the NMJs, aPKC genetic loss of function can dominantly suppress the synaptic hyperplasia exhibited by dFmr13 homozygote nulls (Figures 6G–6J). This result is the reverse of the gain-of-function suppression interaction from overexpression observed in the eye and demonstrates that aPKC and dFmr1 function together in vivo. Table 1. mRNAs Associated with dLgl via dFmr1 wt wt/dfmr13 wt dfmr13 Gene dFmr1 IP/Input dFmr1 IP dLgl IP/Input dLgl IP/Input Biological Process CG16969 CG4101 CG6136 CG9293 CG9681 CG14210 CG3348 CG6444 Cbl 7.6 15.0 18.3 12.0 8.2 3.3 9.1 5.3 4.14 3.5 2.5 2.5 3 2.1 2.3 2.6 2.3 2.3 4.5 6.7 8.1 6.1 5.73 2.5 5.1 3.5 7.7 NC NC NC NC NC NC NC NC NC unknown unknown ion transport apoptosis/cell cycle immunity/circadian rhythm unknown circadian rhythm unknown cell cycle Enrichment is shown in actual numbers (not log scale) as an average from two different experiments. NC, not changed. Biological process as predicted or annotated in Flybase (http://www.flybase.org/). Fmr Protein Acts with Lgl and the PAR Complex 49 We used immunoprecipitation experiments to directly detect an aPKC-zeta/mFmrp/mLgl triple complex in the mouse brain (Figure 6K). aPKC-zeta also associates with dFmr1 and dLgl in flies, albeit more weakly (data not shown). Fmrp function can be regulated by phosphorylation of three serine residues, which are conserved between flies and mammals (Ceman et al., 2003; Siomi et al., 2002). In vertebrates, phosphorylated Fmrp is associated with stalled polyribosomes, and thus phosphorylation is believed to modulate translation regulation of target mRNAs (Ceman et al., 2003). We used a kinase assay to show that aPKC-zeta has the ability to phosphorylate full-length dFmr1 in vitro (Figure 6L). Although dFmr1 possesses four aPKC-zeta phosphorylation consensus sites, when aPKC-zeta is reduced more than 90%, we cannot detect clear mobility shifts in dFmr1 protein (data not shown). It remains formally possible that aPKC-zeta acts a kinase for dFmr1 in vivo, within special developmental or cellular contexts. This is consistent with previous work showing that aPKC-zeta can phosphorylate Lgl as well as Crumbs protein in independent complexes (Betschinger et al., 2003; Sotillos et al., 2004). Taken together, these genetic and biochemical data suggest that the Fmrp/Lgl function may be in part regulated by the interaction with the PAR complex, although the dynamics of the Fmrp/Lgl/PAR complex remains to be established. Discussion Here we report the identification of Lgl as a functional partner of the Fragile X protein, Fmrp. Lgl forms a large macromolecular complex with Fmrp, which is developmentally regulated and modulates the architecture of the neuromuscular junction in the fly. At the cellular level, Lgl and Fmrp are temporally and spatially coexpressed during development and colocalize in granules in the soma and the developing neurites of mouse cultured cells. Fractionation experiments show that Fmrp and Lgl comigrate with Golgi membrane-associated complexes. Furthermore, the Fmrp/Lgl complex contains a subset of mRNAs and interacts physically and genetically with the PAR complex, an essential component of the cellular polarization pathway. Our results suggest that Lgl functions with Fmrp to regulate a subset of target mRNAs during synaptic development and/or function. We propose that Lgl may regulate Fmrp/mRNA containing RNPs by (1) sorting at the Golgi, (2) transport in neurites, and (3) anchoring at specific membrane domains, such Figure 6. Fmrp Interacts with the PAR Complex (A and B) Immunolocalization of dFmr1 and Baz show upregulation in the developing oocyte (see arrows). (C–F) Adult compound eyes, scanning electron micrographs (genotypes as indicated). (G–I) Neuromuscular junctions (NMJs), segment 3, muscle 6/7. aPKC null allele is a significant dominant suppressor of dFmr1 (compare [G], [H], and [I]). (J) Histogram: number of boutons per junction, error bars: standard error of the mean. dFmr13 (102 boutons, n ⫽ 13) and aPKC06403/⫹ (81 boutons, n ⫽ 12) both show elevated bouton numbers relative to wild-type (see Figure 2A), but this elevation is largely rescued in aPKC/⫹; dFmr13 (66 boutons, n ⫽ 8); the difference between dFmr13 and aPKC/⫹; dFmr13 is statistically significant (p ⬍ 0.001, asterisk). (K) aPKC-zeta associates with mFmrp and mLgl in developing mouse brains. Precipitating antibody indicated to the left, blotting antibody to the right. (L) aPKC-zeta phosphorylates dFmr1 in vitro. dFmr1 expressed as fusion with Maltose Binding Protein (MBP). Coomassie-stained gel on top, autoradiograph on the bottom, lanes loaded as indicated. Scale bars equal 20 m in (A), 50 m in (C), and 15 m in (G). Developmental Cell 50 as subsynaptic sites. The neurite transport function may involve molecular motors such as myosin II and kinesin, previously shown to associate with Lgl and Fmrp, respectively (Bilder, 2004; Ohashi et al., 2002). The anchoring mechanism may involve the PAR complex, which has a demonstrated role in defining membrane domains and has recently been shown to generate asymmetry in the C. elegans embryo by stabilizing RNPs at the posterior pole (Cheeks et al., 2004). Lgl Interacts Genetically and Forms a Functional Complex with Fmrp in Neural Development Our data demonstrate that Fmrp and Lgl form a functional complex in living neurons, and this is conserved in flies and mice. In the mouse brain, mFmrp associates with mLgl preferentially at a time of increased synaptogenesis, demonstrating a developmentally regulated interaction between Lgl and Fmrp. Our data suggest that dLgl acts to regulate a subset of dFmr1-associated mRNAs with some encoding circadian regulated molecules (CG3348 and CG9681) and some encoding secreted or transmembrane proteins (CG6136, CG4101, CG9681) among others. dLgl also associates with mRNA independent of dFmr1 (see Supplemental Data); thus, it is formally possible that dLgl interacts with other RNA binding proteins, which remain to be determined. Fmrp has been implicated in the translational regulation of specific target mRNAs, perhaps via the RNAi pathway (Carthew, 2002; Jin et al., 2004). Also, it has been proposed that Fmrp is involved in the transport and localization of mRNAs: cellular fractionation and immunolocalization data revealed the association of Fmrp-containing complexes with molecular motors such as kinesin and myosin V (De Diego Otero et al., 2002; Ohashi et al., 2002). Lgl functions in cellular polarity via regulating myosin motor activity and/or vesicle transport (Bilder, 2001) and has been shown to regulate polarized delivery by sorting at the Golgi (Bilder, 2004). Taken together, these concepts and our data suggest that Lgl may act as a scaffold for Fmrp granules, possibly at the Golgi, and perhaps aids in carrying specific mRNA targets to sites of locally controlled translation (such as subsynaptic sites). A Fmrp/Lgl/PAR Complex May Function as a Synaptic Tag in Neural Development It is also possible that Lgl anchors Fmrp at specific membrane domains such as synapses, perhaps via the PAR complex. Lgl function is regulated by the PAR complex, specifically via phosphorylation by aPKC-zeta and by direct binding to PAR6 (Betschinger et al., 2003; Plant et al., 2003). Our genetic interaction data suggest that PAR6 and Baz antagonize dFmr1 function, which is in accordance with previously published work showing that the PAR complex inhibits dlgl (Betschinger et al., 2003) and with our data that demonstrate that dlgl functions cooperatively with dFmr1. aPKC-zeta was shown to antagonize most dLgl functions with the exception of its role in regulating neuroblast apical size (Bilder, 2004), and we find our data to be consistent with such reports. In our hands, loss of function for aPKC-zeta suppresses gain of function sev:dFmr1 as well as the loss-of-function phenotype of dFmr1 at the NMJs. Taken together, these data suggest a dynamic relationship between the various members of the complex. One possible interpretation of our results is that aPKC/PAR can act on Fmrp directly or via Lgl depending on the developmental and/or cellular context. Furthermore, this is consistent with our sucrose fractionation experiments, which suggest the existence of at least two complexes comprising dFmr1/PAR proteins and dFmr1/ dLgl/PAR proteins. The PAR complex not only functions in cell polarity, but also at the synapse, where it is believed to function in synaptic tagging (Martin and Kosik, 2002). Synaptic tags have been proposed to transiently mark a synapse after activation in a way that will translate the local events into persistent functional changes (such as longterm depression), processes in which Fmrp is also thought to act (Huber et al., 2002). Thus, the Fmrp/Lgl/ PAR complex may act in synaptic plasticity linking synaptic input to the remodeling of the cytoskeleton and mediating required translational changes. Experimental Procedures Nomenclature Naming conventions used here are as follows: (1) human: FMR1 (gene), FMRP (protein); (2) fly: dFmr1 (for Fmr1 gene, also known as dfxr), dFmr1 (protein), dlgl (for l(2)gl gene), dLgl (protein); (3) mouse: mFmrp (protein), mLgl (protein); (4) generic: Fmrp (protein), Lgl (protein). Drosophila Stocks, Genetic Screen, and Mapping Genotypes: w1118; cn1; P(w, ry)E)7(94D), w1118; l(2)DTS911 nocSco/ sev:dFmr1 CyO, w1118; dlgl4/kr:GFP CyO; dFmr13/TM6B, w1118; UASdFmr1 UAS-dlgl/TM6B, baz4/FM7i, par6⌬226/FM7i, daPKC06403/kr:GFP CyO. dlgl overexpression: UAS-dlgl/⫹; dFmr1 driven by: BG 487 (postsynaptic), elav-Gal4 (presynaptic). Genetic screen: isogenic w1118; cn1; (P(w, ry)E)7, 25 mM EMS (Sigma), treated males crossed to autosomally isogenic w1118; l(2)DTS911 nocSco/sev:dFmr1 CyO. New dlgl alleles named: dlgla1 - dlgla19, a single lethal complementation group, maps left of P(w⫹mC⫽lacW)l(2)k11324 (by meiotic recombination), left of EP(2)456 (by P-induced male recombination), fails to complement Df(2L)TE21A, dlgl334 and dlgl275. Lethality rescued by dlgl transgene (Ins(2R) p(l(2)gl⫹ neo30)53DE); Mechler et al., 1985). Temperature shift for oocyte morphology: wild-type and lglts3 shifted to 29⬚C for 24. Microscopy and Immunohistochemistry Statistics as previously described (Kumar et al., 2003). Wild-type: mean rhabdomeres/facet ⫽ 6.99, standard deviation (SD) ⫽ 0.07 (188 ommatidia counted in 3 eyes), sev: dFmr1/⫹: mean ⫽ 6.06, SD ⫽ 0.13 (331, in 3 eyes), sev: dFmr1/dlgl334 mean ⫽ 6.60, SD ⫽ 0.14 (281, in 3 eyes). Polyclonal anti-Lgl antibodies as described (Ohshiro et al., 2000). Quantitative analysis in Figure 1 by Photoshop (Adobe): mean and SD of signal for three 50 square pixel areas anterior to the furrow (endogenous), compared to three 100 pixel square areas, posterior (ectopic). Reduction of dlgl dosage: 22% decrease in signal (149.3 ⫾ 2.9 in Figure 1J, reduced to 116.3 ⫾ 2.4 in Figure 1K). Anti-dFmr1 6A15, anti-mLgl, anti-CSP, anti-mFmrp, anti-Baz antibodies as described (Plant et al., 2003; Wan et al., 2000; Wodarz et al., 2000; Zinsmaier et al., 1994). Rhodamine-, FITC-, and Cy5-conjugated secondary antibodies (Jackson ImmunoResearch). Eye discs, ovaries, and NMJs prepared as described (Kumar et al., 2003; Rohrbaugh et al., 2000; Wodarz et al., 2000; Zarnescu and Thomas, 1999). Analysis of boutons adapted from Rohrbaugh et al. (2000). Surface areas of muscles 6/7 were measured by Photoshop. Cloning, IP, and Pull-Down Experiments IP: w1118 flies frozen in liquid nitrogen, heads collected, homogenization and processing adapted from Brown et al. (2001). Final washes in 50 mM NaCl, 50 mM Tris (pH 7.5). Pull-down: FLAG-dFmr1 (TnT, Fmr Protein Acts with Lgl and the PAR Complex 51 Promega), incubated with M2 Sepharose (Sigma) and fly extract, nonspecific interactions competed with FLAG peptide (Sigma). Mouse brain IPs as described (Brown et al., 2001). SmaI-XbaI dFmr1 cDNA fragment inserted into pCMF (Invitrogen). CFP-mlgl: gift from T. Pawson (University of Toronto). References Tissue Culture MEFs from Immortomouse (Charles River Laboratory; P.J. and S.T.W., unpublished), in DMEM ⫹10% FCS, antibiotics, ␥-interferon at 33⬚C, replated without interferon, incubated at 37⬚C overnight, then transfected with pCMV-RFP-mFMRP or pCMV-CFP-mLgl1 (Lipofectamine 2000, Invitrogen). After 36 hr, cells stained for mLgl (Plant et al., 2003). 97.4% of RFP-mFmrp transfected cells show endogenous mLgl redistribution (n ⫽ 80, 38 transfected). CAD cells transfected similarly and induced after 24 hr. Betschinger, J., Mechtler, K., and Knoblich, J.A. (2003). The Par complex directs asymmetric cell division by phosphorylating the cytoskeletal protein Lgl. Nature 422, 326–330. RNA Isolation and Oligonucleotide Array Expression Analysis RNA from precleared lysate and IPed complexes isolated using Trizol (GIBCO-BRL), cleaned with the RNeasy (QIAGEN). cRNA generated per Affymetrix Manual, protocols P/N 900218rev.2, P/N700219rev.3. Absolute and Comparison gene chip analyses with Affymetrix Microarray Suite 4.0. Each experiment done 2–3 times and two chips probed each time (see Supplemental Data and NCBI GEO GPL1522). Data filtered for probe sets “present” and “increased” in WT-IP RNA with a ⫹1.4-fold change or greater. Arquier, N., Perrin, L., Manfruelli, P., and Semeriva, M. (2001). The Drosophila tumor suppressor gene lethal(2)giant larvae is required for the emission of the Decapentaplegic signal. Development 128, 2209–2220. Bilder, D. (2001). Cell polarity: squaring the circle. Curr. Biol. 11, R132–R135. Bilder, D. (2004). Epithelial polarity and proliferation control: links from the Drosophila neoplastic tumor suppressors. Genes Dev. 18, 1909–1925. Bilder, D., Li, M., and Perrimon, N. (2000). Cooperative regulation of cell polarity and growth by Drosophila tumor suppressors. Science 289, 113–116. Brown, V., Jin, P., Ceman, S., Darnell, J.C., O’Donnell, W.T., Tenenbaum, S.A., Jin, X., Feng, Y., Wilkinson, K.D., Keene, J.D., et al. (2001). Microarray identification of FMRP-associated brain mRNAs and altered mRNA translational profiles in fragile X syndrome. Cell 107, 477–487. Carthew, R.W. (2002). RNA interference: the fragile X syndrome connection. Curr. Biol. 12, R852–R854. Ceman, S., O’Donnell, W.T., Reed, M., Patton, S., Pohl, J., and Warren, S.T. (2003). Phosphorylation influences the translation state of FMRP-associated polyribosomes. Hum. Mol. Genet. 12, 3295–3305. Cellular and Membrane Fractionation and Western Blots Larval extract spun at 200K ⫻ g, 2–3 mg total protein layered onto 5%–20% sucrose gradients, processed as in Salazar et al. (2004). Membrane fractions: extracts from 2.5- to 3.5-hr-old Oregon R embryos as in Sisson et al. (2000). Homogenate spun 3K ⫻ g, 10 min, 4⬚C, supernatant (S3) spun at 10K ⫻ g, 15 min, 4⬚C. Supernatant (S10) onto sucrose cushion (15 mM Tris [pH 7.0], 50 mM KCl, 0.5 mM EDTA, and 0.5 M sucrose), spun at 100K ⫻ g, 1 hr, 4⬚C. Pellet (P100) washed in 15 mM Tris (pH 7.0), 50 mM KCl, and 0.5 mM EDTA, resuspended and mixed with Optiprep (Accurate Chemical & Scientific Corp.) for a 10%–30% density gradient, spun at 340K ⫻ g, 3 hr, 4⬚C. 0.25 ml fractions to 6.5%–12% SDS-PAGE and blotted (Sisson et al., 2000). Westerns: rabbit anti-dLgl (1:10,000), mouse anti-dFmr1 (1:10,000), rabbit anti-Lva (1:10,000), rat anti-BiP (1:100 Affinity Bioreagents Inc.), rabbit anti-aPKC-zeta (1:1000 Sigma), rabbit anti-PAR6 (1:2000, (Betschinger et al., 2003), HRP secondary antibodies (1:5,000–10,000, Santa Cruz Biotechnology). Cox, D.N., Seyfried, S.A., Jan, L.Y., and Jan, Y.N. (2001). Bazooka and atypical protein kinase C are required to regulate oocyte differentiation in the Drosophila ovary. Proc. Natl. Acad. Sci. USA 98, 14475–14480. In Vitro Kinase Assay MBP-dFMR1 into pMAL-c2x (NEB). MBP and MBP-dFMR1 expressed and purified. In vitro kinase assays as in Betschinger et al. (2003). De Lorenzo, C., Strand, D., and Mechler, B.M. (1999). Requirement of Drosophila I(2)gl function for survival of the germline cells and organization of the follicle cells in a columnar epithelium during oogenesis. Int. J. Dev. Biol. 43, 207–217. Acknowledgments Gao, F.B. (2002). Understanding fragile X syndrome: insights from retarded flies. Neuron 34, 859–862. We thank the Bloomington Drosophila stock center; V. Budnik, C. Doe, J. Fawcet, F. Matsuzaki, T. Pawson, K. Zinsmaier, A. Wodarz, and especially B. Mechler for reagents; R. Ordway, S. Ceman, and especially V. Faundez for technical advice; and S. Cook, R. Airinei, D. Saiman, M. Dawson, W. Pierre, and the Emory IM&MF for technical assistance. We thank K. Moberg and members of the Moses lab for comments on the manuscript. This paper is dedicated to the memory of M. Diaconu. D.C.Z. and T.C.D. were supported by fellowships from FRAXA. T.C.D. and T.A.J. were supported by NIH HD33834. P.J. is supported by Rett Syndrome Research Foundation, Y.F. by NIH 5 PO1 HD355765, and S.T.W. by NIH R37HD20521, P01HD35576, and P30HD024064. J.C.S. supported by a March of Dimes Basil O’Connor Award and RO1 GM067013 from the NIH. J.A.K.’s lab supported by Boehringer Ingelheim, the Austrian Academy of Sciences, and the Wiener Wirtschaftsfoerderungsfonds. D.C.Z. is supported by NIH F32 NS046880. This work was chiefly supported by NIH R21 NS43536. Received: June 23, 2004 Revised: September 19, 2004 Accepted: October 22, 2004 Published: January 3, 2005 Cheeks, R.J., Canman, J.C., Gabriel, W.N., Meyer, N., Strome, S., and Goldstein, B. (2004). C. elegans PAR proteins function by mobilizing and stabilizing asymmetrically localized protein complexes. Curr. Biol. 14, 851–862. Darnell, J.C., Jensen, K.B., Jin, P., Brown, V., Warren, S.T., and Darnell, R.B. (2001). Fragile X mental retardation protein targets G quartet mRNAs important for neuronal function. Cell 107, 489–499. De Diego Otero, Y., Severijnen, L.A., van Cappellen, G., Schrier, M., Oostra, B., and Willemsen, R. (2002). Transport of fragile X mental retardation protein via granules in neurites of PC12 cells. Mol. Cell. Biol. 22, 8332–8341. Hinton, V.J., Brown, W.T., Wisniewski, K., and Rudelli, R.D. (1991). Analysis of neocortex in three males with the fragile X syndrome. Am. J. Med. Genet. 41, 289–294. Huber, K.M., Gallagher, S.M., Warren, S.T., and Bear, M.F. (2002). Altered synaptic plasticity in a mouse model of fragile X mental retardation. Proc. Natl. Acad. Sci. USA 99, 7746–7750. Jakobs, R., de Lorenzo, C., Spiess, E., Strand, D., and Mechler, B.M. (1996). Homo-oligomerization domains in the lethal(2)giant larvae tumor suppressor protein, p127 of Drosophila. J. Mol. Biol. 264, 484–496. Jin, P., Zarnescu, D.C., Ceman, S., Nakamoto, M., Mowrey, J., Jongens, T.A., Nelson, D.L., Moses, K., and Warren, S.T. (2004). Biochemical and genetic interaction between the fragile X mental retardation protein and the microRNA pathway. Nat. Neurosci. 7, 113–117. Klezovitch, O., Fernandez, T.E., Tapscott, S.J., and Vasioukhin, V. (2004). Loss of cell polarity causes severe brain dysplasia in Lgl1 knockout mice. Genes Dev. 18, 559–571. Knust, E., and Bossinger, O. (2002). Composition and formation of intercellular junctions in epithelial cells. Science 298, 1955–1999. Kumar, J.P., Hsiung, F., Powers, M., and Moses, K. (2003). Nuclear Developmental Cell 52 translocation of activated MAP kinase is developmentally regulated in the developing Drosophila eye. Development 130, 3703–3714. Larsson, K., Bohl, F., Sjostrom, I., Akhtar, N., Strand, D., Mechler, B.M., Grabowski, R., and Adler, L. (1998). The Saccharomyces cerevisiae SOP1 and SOP2 genes, which act in cation homeostasis, can be functionally substituted by the Drosophila lethal(2)giant larvae tumor suppressor gene. J. Biol. Chem. 273, 33610–33618. Lee, A., Li, W., Xu, K., Bogert, B.A., Su, K., and Gao, F.B. (2003). Control of dendritic development by the Drosophila fragile X-related gene involves the small GTPase Rac1. Development 130, 5543– 5552. Schenck, A., Bardoni, B., Langmann, C., Harden, N., Mandel, J.L., and Giangrande, A. (2003). CYFIP/Sra-1 controls neuronal connectivity in Drosophila and links the Rac1 GTPase pathway to the fragile X protein. Neuron 38, 887–898. Siomi, M.C., Higashijima, K., Ishizuka, A., and Siomi, H. (2002). Casein kinase II phosphorylates the fragile X mental retardation protein and modulates its biological properties. Mol. Cell. Biol. 22, 8438– 8447. Sisson, J.C., Field, C., Ventura, R., Royou, A., and Sullivan, W. (2000). Lava lamp, a novel peripheral golgi protein, is required for Drosophila melanogaster cellularization. J. Cell Biol. 151, 905–918. Lehman, K., Rossi, G., Adamo, J.E., and Brennwald, P. (1999). Yeast homologues of tomosyn and lethal giant larvae function in exocytosis and are associated with the plasma membrane SNARE, Sec9. J. Cell Biol. 146, 125–140. Sotillos, S., Diaz-Meco, M.T., Caminero, E., Moscat, J., and Campuzano, S. (2004). DaPKC-dependent phosphorylation of Crumbs is required for epithelial cell polarity in Drosophila. J. Cell Biol. 166, 549–557. Manfruelli, P., Arquier, N., Hanratty, W.P., and Semeriva, M. (1996). The tumor suppressor gene, lethal(2)giant larvae (1(2)g1), is required for cell shape change of epithelial cells during Drosophila development. Development 122, 2283–2294. Strand, D., Jakobs, R., Merdes, G., Neumann, B., Kalmes, A., Heid, H.W., Husmann, I., and Mechler, B.M. (1994a). The Drosophila lethal(2)giant larvae tumor suppressor protein forms homo-oligomers and is associated with nonmuscle myosin II heavy chain. J. Cell Biol. 127, 1361–1373. Martin, K.C., and Kosik, K.S. (2002). Synaptic tagging—who’s it? Nat. Rev. Neurosci. 3, 813–820. Mazroui, R., Huot, M.E., Tremblay, S., Filion, C., Labelle, Y., and Khandjian, E.W. (2002). Trapping of messenger RNA by fragile X mental retardation protein into cytoplasmic granules induces translation repression. Hum. Mol. Genet. 11, 3007–3017. McDonald, M.J., and Rosbash, M. (2001). Microarray analysis and organization of circadian gene expression in Drosophila. Cell 107, 567–578. Mechler, B.M., McGinnis, W., and Gehring, W.J. (1985). Molecular cloning of lethal(2)giant larvae, a recessive oncogene of Drosophila melanogaster. EMBO J. 4, 1551–1557. Nimchinsky, E.A., Oberlander, A.M., and Svoboda, K. (2001). Abnormal development of dendritic spines in fmr1 knock-out mice. J. Neurosci. 21, 5139–5146. O’Donnell, W.T., and Warren, S.T. (2002). A decade of molecular studies of fragile X syndrome. Annu. Rev. Neurosci. 25, 315–338. Ohashi, S., Koike, K., Omori, A., Ichinose, S., Ohara, S., Kobayashi, S., Sato, T.A., and Anzai, K. (2002). Identification of mRNA/protein (mRNP) complexes containing Pur-alpha, mStaufen, fragile X protein, and Myosin Va and their association with rough endoplasmic reticulum equipped with a kinesin motor. J. Biol. Chem. 277, 37804– 37810. Ohno, S. (2001). Intercellular junctions and cellular polarity: the PARaPKC complex, a conserved core cassette playing fundamental roles in cell polarity. Curr. Opin. Cell Biol. 13, 641–648. Ohshiro, T., Yagami, T., Zhang, C., and Matsuzaki, F. (2000). Role of cortical tumour-suppressor proteins in asymmetric division of Drosophila neuroblast. Nature 408, 593–596. Peng, C.Y., Manning, L., Albertson, R., and Doe, C.Q. (2000). The tumour-suppressor genes lgl and dlg regulate basal protein targeting in Drosophila neuroblasts. Nature 408, 596–600. Plant, P.J., Fawcett, J.P., Lin, D.C., Holdorf, A.D., Binns, K., Kulkarni, S., and Pawson, T. (2003). A polarity complex of mPar-6 and atypical PKC binds, phosphorylates and regulates mammalian Lgl. Nat. Cell Biol. 5, 301–308. Qi, Y., Wang, J.K., McMillian, M., and Chikaraishi, D.M. (1997). Characterization of a CNS cell line, CAD, in which morphological differentiation is initiated by serum deprivation. J. Neurosci. 17, 1217–1225. Rohrbaugh, J., Grotewiel, M.S., Davis, R.L., and Broadie, K. (2000). Integrin-mediated regulation of synaptic morphology, transmission and plasticity. J. Neurosci. 20, 6868–6878. Ruiz-Canada, C., Ashley, J., Moeckel-Cole, S., Drier, E., Yin, J., and Budnik, V. (2004). New synaptic bouton formation is disrupted by misregulation of microtubule stability in aPKC mutants. Neuron 42, 567–580. Salazar, G., Love, R., Werner, E., Doucette, M.M., Cheng, S., Levey, A., and Faundez, V. (2004). The zinc transporter ZnT3 interacts with AP-3 and it is preferentially targeted to a distinct synaptic vesicle subpopulation. Mol. Biol. Cell 15, 575–587. Strand, D., Raska, I., and Mechler, B.M. (1994b). The Drosophila lethal(2)giant larvae tumor suppressor protein is a component of the cytoskeleton. J. Cell Biol. 127, 1345–1360. Tepass, U., Tanentzapf, G., Ward, R., and Fehon, R. (2001). Epithelial cell polarity and cell junctions in Drosophila. Annu. Rev. Genet. 35, 747–784. Wan, L., Dockendorff, T.C., Jongens, T.A., and Dreyfuss, G. (2000). Characterization of dFMR1, a Drosophila melanogaster homolog of the fragile X mental retardation protein. Mol. Cell. Biol. 20, 8536– 8547. Wodarz, A., Ramrath, A., Grimm, A., and Knust, E. (2000). Drosophila atypical protein kinase C associates with Bazooka and controls polarity of epithelia and neuroblasts. J. Cell Biol. 150, 1361–1374. Zarnescu, D.C., and Thomas, G.H. (1999). Apical spectrin is essential for epithelial morphogenesis but not apicobasal polarity in Drosophila. J. Cell Biol. 146, 1075–1086. Zhang, Y.Q., Bailey, A.M., Matthies, H.J., Renden, R.B., Smith, M.A., Speese, S.D., Rubin, G.M., and Broadie, K. (2001). Drosophila fragile X-related gene regulates the MAP1B homolog Futsch to control synaptic structure and function. Cell 107, 591–603. Zinsmaier, K.E., Eberle, K.K., Buchner, E., Walter, N., and Benzer, S. (1994). Paralysis and early death in cysteine string protein mutants of Drosophila. Science 263, 977–980.