Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Dev Biol. Author manuscript; available in PMC 2013 May 3.
Published in final edited form as:
PMCID: PMC3642870

Combinatorial use of translational co-factors for cell type-specific regulation during neuronal morphogenesis in Drosophila


The translational regulators Nanos (Nos) and Pumilio (Pum) work together to regulate the morphogenesis of dendritic arborization (da) neurons of the Drosophila larval peripheral nervous system. In contrast, Nos and Pum function in opposition to one another in the neuromuscular junction to regulate the morphogenesis and the electrophysiological properties of synaptic boutons. Neither the cellular functions of Nos and Pum nor their regulatory targets in neuronal morphogenesis are known. Here we show that Nos and Pum are required to maintain the dendritic complexity of da neurons during larval growth by promoting the outgrowth of new dendritic branches and the stabilization of existing dendritic branches, in part by regulating the expression of cut and head involution defective. Through an RNA interference screen we uncover a role for the translational co-factor Brain Tumor (Brat) in dendrite morphogenesis of da neurons and demonstrate that Nos, Pum, and Brat interact genetically to regulate dendrite morphogenesis. In the neuromuscular junction, Brat function is most likely specific for Pum in the presynaptic regulation of bouton morphogenesis. Our results reveal how the combinatorial use of co-regulators like Nos, Pum and Brat can diversify their roles in post-transcriptional regulation of gene expression for neuronal morphogenesis.

Keywords: Brain Tumor, Nanos, Pumilio, dendrite morphogenesis, neuromuscular junction


Neurons are highly polarized cells whose dendritic and axonal processes adopt distinct morphologies necessary for reception and transmission of signals. For example, the specific arborization patterns of dendrites in different types of sensory neurons are essential for neurons to establish their receptive fields and to respond to appropriate signals. At the neuromuscular junction (NMJ), axons must form specialized synaptic structures called boutons to regulate muscle dynamics. Post-transcriptional regulatory mechanisms including mRNA splicing, mRNA transport, and local control of protein synthesis have been implicated in the development of such polarized morphologies (Baines, 2005; Wu et al. 2007) and mutations in RNA-binding proteins involved in these processes have been linked to neuronal morphology defects associated with neurodegenerative and mental retardation disorders (Bassell and Warren, 2008; Fallini et al. 2011).

The dendritic arborization (da) neurons in the Drosophila larval peripheral nervous system (PNS) have provided a model for studying mechanisms underlying dendrite morphogenesis, including post-transcriptional control. These neurons are divided into four classes based on the complexity of their dendritic arbors, with class IV neurons exhibiting the most highly complex branching patterns. During late embryogenesis and early larval development, class IV da neurons elaborate their dendrites in a non-overlapping manner to establish large receptive fields that cover the larval body wall. Subsequently, as the larva increases dramatically in size, dendrites grow in synchrony with the body wall epithelium to maintain receptive field coverage (Parrish et al., 2009). Finally, class IV da neuron dendrites are pruned during pupariation in a process involving branch severing and caspase-mediated degeneration and are replaced by dendrites appropriate for the adult nervous system (Williams and Truman, 2005a; Williams and Truman, 2005b).

Mutations in the evolutionarily conserved translational repressors Nanos (Nos) and Pumilio (Pum) result in reduced higher order dendritic branching and dendritic field coverage in class IV da neurons (Brechbiel and Gavis, 2008; Ye et al. 2004). Nos and Pum were first identified as components of a translational repression complex that regulates embryonic patterning (Parisi and Lin, 2000), and genetic analysis indicates that they also act collaboratively in class IV da neurons (Ye et al. 2004). Moreover, dendritic localization of nos mRNA is required for dendrite branching morphogenesis (Brechbiel and Gavis, 2008), suggesting that Nos and Pum might mediate local control of translation.

Pum binds specifically to a sequence termed the Nanos Response Element (NRE) in the 3′ untranslated region (3′UTR) of target mRNAs. In a paradigm derived from analysis of Nos/Pum-dependent regulation of hunchback (hb) mRNA in the early embryo, Pum recruits Nos to the NRE to form a ternary complex, to which a third protein, Brain tumor (Brat) is then recruited to form a functional repression complex (Muraro et al., 2008; Sonoda and Wharton, 2001). Regulation of some targets by the Nos/Pum complex, however, does not require Brat (Harris et al., 2011; Muraro et al. 2008). Moreover, Nos and Pum can act independently and even in opposition to one another. In the ovary, Pum functions independently of Nos, but together with Brat to promote germline stem cell differentiation (Harris et al., 2011). In the NMJ, Nos and Pum have opposing functions in regulating both bouton formation and glutamate receptor subunit composition, thereby differentially regulating NMJ physiology (Menon et al., 2009). Thus, Nos, Pum, and Brat participate combinatorially in distinct complexes that act on different targets. Whether Brat is required by Nos and/or Pum in da neurons and in the NMJ is unknown.

To better understand the role of translational control in neuronal morphogenesis, we investigated the specific functions of Nos and Pum in dendrite elaboration by sensory neurons. Live imaging analysis revealed that Nos and Pum are required to maintain dendritic complexity during the third larval instar by promoting the addition of new branches and the stabilization of existing branches. We show that Nos and Pum control dendritic branching in part by repressing the expression of the pro-apoptotic factor Head Involution Defective (Hid). In addition, Nos and Pum exert differential effects on the level of the transcription factor Cut, indicating distinct target-specific requirements for these translational repressors. Finally, we uncover a role for Brat in the development of class IV da neurons and show that Brat functions together with Nos and Pum in dendrite morphogenesis. In contrast, Brat functions similarly to Pum and in opposition to Nos to regulate bouton morphogenesis in the NMJ. Our results show how control of gene expression during development can be diversified through the combinatorial use of translational co-regulators to generate distinct neuronal morphologies and, potentially, functions.

Materials and Methods

Fly Strains

The following fly strains were used: GAL4477, UAS-mCD8:GFP (Grueber et al., 2003a); ppk-GAL4, UAS-mCD8:GFP (Grueber et al., 2007), OK6-GAL4, UAS-syt:GFP (Aberle et al., 2002); Mef2-GAL4 (Ranganayakulu et al., 1996); MHC82-Gal4 (Marek et al., 2000); UAS-nosRNAi (Menon et al., 2009); UAS-bratRNAiVDRC (P{KK113206}; Dietzl et al., 2007); UAS-bratRNAiTRIP (P{TRiP.HMS01121}; Transgenic RNAi Project, Harvard Medical School); UAS-brat (Sonoda and Wharton, 2001); UAS-bratRD (Harris et al., 2011); UAS-nos-tub3′UTR (Clark et al., 2002); UAS-pum (Ye et al., 2004); dlg::YFP (Rees et al., 2011); Df(3L)H99 (Abbott and Lengyel, 1991); ctc145 (Grueber et al., 2003); d4EHPCP53 (Cho et al., 2006). Larvae mutant for nos were generated by using the strong hypomorphic combination nosRC/nosRD (Curtis et al., 1997). Larvae mutant for brat were generated using brat11/Df(2L)TE37C-7 (Frank et al., 2002; Sonoda and Wharton, 2001) or the strong hypomorphic combination brat1/brat11 (Sonoda and Wharton, 2001). Larvae mutant for pum were generated using the strong hypomorphic combination pumET7/pumET9 (Forbes and Lehmann, 1998). For RNAi and overexpression experiments, UAS transgenes were used in single copy. Animals were maintained at 25°C except in experiments using ppk-GAL4 and in hid epistasis experiments, which were performed at 29°C to increase GAL4 efficiency.

For MARCM, FRT40 brat11 flies (Frank et al., 2002) were mated to elav-GAL4, UAS-mCD8:GFP, hs-FLP; FRT40A tubP-GAL80 flies (Lee and Luo, 2001). Embryos were collected for a 2 hour period and aged for 3 hours at 25°C. Embryos were then heat-shocked at 39°C for 50 minutes, allowed to recover for 30 minutes at 25°C, then heat-shocked again at 39°C for 45 minutes. Animals were reared at 18°C until the late third (wandering) larval stage, when GFP-positive clones were imaged.

Imaging and quantification of dendrite and bouton morphology

Live imaging analysis was performed as previously described (Parrish et al., 2007a) except that larvae were mounted in halocarbon oil and imaged on a Leica SPE confocal microscope using a 40x/1.25 NA oil objective. The total number of terminal branches, mean branch length, terminal branch length, as well as the number of branches lost or gained between two time points were quantified in Z series projections of a single ddaC neuron by manual counting and by analyzing tracings of neurons created with NeuronJ (Meijering et al., 2004). All dendritic termini visible in the field of view were analyzed.

Unless otherwise noted, dendrite and bouton morphology were examined in wandering larval stages, corresponding to approximately 108–120 hours after egg laying at 25°C. Morphology was analyzed in larval fillet preparations (Ye et al., 2004) immunostained with 1:350 Alexa Fluor 488 rabbit anti-GFP (Invitrogen), mounted in 70% glycerol, and imaged on a Leica SP5 confocal microscope using a 40x/1.25 NA oil objective (da neurons) or the 40x objective with 1.5x zoom (NMJ boutons). The total number of terminal branches was quantified in projections of individual ddaC neurons from the second through fifth abdominal segment as previously described (Lee et al., 2003). For wild-type ddaC neurons visualized using either GAL4477 or ppk-GAL4 to express mCD8-GFP, approximately 350 terminal branches are routinely detected within the field of view. For bouton analysis, the NMJ lying at the interface of muscle 6/7 within the second and third abdominal segment was imaged. The total number of synaptic boutons per NMJ was determined separately for the second and third abdominal segments. Statistical significance was determined using the Student’s t-test.


Wandering larvae (approximately 108–120 hours after egg laying at 25°C) were filleted, fixed in 4% EM grade formaldehyde (Polysciences), and incubated overnight at 4°C with the following primary antibodies in PBS/0.1% Triton X-100/5% normal goat serum (NGS): 1:20 mouse anti-cut (Developmental Studies Hybridoma Bank [DSHB]); 1:500 mouse anti-Futsch (DSHB); 1:10 mouse anti-Syt (DSHB); 1:100 mouse anti-Dlg (DSHB); 1:350 rat anti-Brat333 (Sonoda and Wharton, 2001); 1:200 rabbit anti-Brat (Cho et al., 2006); 1:100 mouse anti-Hid (gift from H. Steller); 1:100 rabbit anti-activated caspase (Trevigen). The following secondary antibodies (Invitrogen) were incubated overnight at 4°C in PBS/0.1% Triton X-100/5% NGS: 1:700 Alexa Fluor 568 goat anti-mouse, 1:500 Alexa Fluor 488 goat anti-mouse; 1:1000 Alexa Fluor 546 goat anti-rat; 1:500 Alexa 568 goat anti-rabbit. For each experiment, all neurons were imaged at the same settings on a Leica SPE confocal microscope with a 63x/1.4 NA oil objective. For anti-Hid and anti-Cut immunostaining, fluorescence intensity was measured within a region of interest (ROI) corresponding to the cell nucleus and normalized to the volume of the cell nucleus using Volocity software. Values reported represent the average of 15–33 neurons from each genotype relative to a comparable number of neurons from the corresponding wild-type control. Statistical significance was measured by performing the Student’s t-test.


nos and pum are required for elaboration and stabilization of dendritic branches during late larval growth

The reduction in dendrite branching complexity in nos mutant class IV da neurons is first apparent during the third larval instar, indicating that nos is not required for the initial elaboration of the dendritic tree (Brechbiel and Gavis, 2008). To determine what role nos plays in dendrite morphogenesis, we first performed a time course analysis of dendrite morphology throughout the third larval instar in fixed larval preparations. The dorsal-most class IV da (ddaC) neurons from wild-type and nos mutant larvae were visualized with the mCD8-GFP membrane marker expressed using the GAL4/UAS system with the class IV da neuron-specific GAL4477 driver. In addition, we analyzed ddaC neurons in which nos expression was knocked down by using GAL4477 to express a UAS-nosRNAi transgene. At 72 hours after egg laying (AEL), corresponding to the early third larval instar, nos mutant neurons and nosRNAi neurons are indistinguishable from control neurons. However, at 96 hours AEL, or mid third larval instar, a few da neurons in each nos mutant or nosRNAi larva show reduced higher order branching as monitored by the quantification of branch termini. This phenotype increases in severity and is exhibited by more neurons per larva as larval development progresses (data not shown; Brechbiel and Gavis, 2008). These results suggest that nos is required for the extension of new branches during larval growth, the maintenance of existing branches, or both.

To distinguish among these possibilities, we monitored dendrite dynamics of individual ddaC neurons by imaging live larvae first at 92 hours AEL and then 17 hours later, at 109 hours AEL. This time window encompasses the onset of the nos mutant phenotype and provides sufficient opportunity to observe branch extension and retraction events. In preliminary experiments, we noticed that genetically nos mutant larvae pupate prior to 109 hours AEL. This early pupariation defect in nos mutant larvae may reflect asynchronous development from wild- type larvae. To ensure that our live imaging analysis accurately compared developmental time points between nos deficient and control neurons, nos was depleted specifically in class IV da neurons by GAL4477-mediated expression of UAS-nosRNAi. Consistent with the time course analysis in fixed tissue, live imaging at 92 hours AEL shows no difference in the number of dendritic termini in nos deficient neurons compared to wild-type neurons. At 109 hours AEL, however, nos RNAi results in fewer dendritic branches as compared to wild-type neurons.

To determine whether the apparent reduction in branching reflects loss of previously established branches, we monitored the net change in dendrite number in ddaC neurons between the two time points. Quantification shows that there is a net increase in the number of dendritic termini in wild-type neurons (Fig. 1A,B,G), while there is net decrease in nosRNAi neurons (Fig. 1C,D,G). Since Nos functions with Pum to control dendrite development (Ye et al., 2004), we investigated whether pum mutant neurons exhibit similar dendrite dynamics to nos deficient neurons. Live imaging reveals that pum mutant neurons also exhibit a net loss of dendritic branches over 17 hours (Fig. 1E,F,G). The extent of this loss exceeds that observed in nos deficient neurons, most likely due to the incomplete knockdown of nos by RNAi. The net decrease in dendritic branch termini indicates that both nos and pum play a role in maintaining existing branches during the late third larval instar.

Fig. 1
nos and pum are required for dendritic branch stabilization. (A-F) Confocal z series projections of class IV da neurons marked using GAL4477 to drive expression of UAS-mCD8:GFP. (A, B) A ddaC neuron from a wild-type (WT) larva was imaged at 92 hours AEL ...

To determine whether Nos and Pum are also required for the growth of new dendritic branches during the late third larval instar, we monitored the dynamics of individual dendrites within each neuron. We observed retraction of numerous branches in nosRNAi and pum mutant neurons within the 17 hour time period, whereas little or no retraction was observed in wild-type neurons. Although wild-type control neurons add many new branches during this time, nosRNAi neurons add significantly fewer branches. A strong trend toward less new growth was observed in pum mutants as well (Fig. 1H). Therefore Nos and Pum are required for stabilization of existing dendritic branches as well as for outgrowth of new branches during the mid to late third larval instar.

Hid is upregulated in nos and pum mutant neurons

Upon puparium formation, class IV da neurons are extensively pruned and remodeled. During this process, primary branches undergo severing and local degeneration, and are then replaced with new dendritic branches necessary for establishing connections in the adult animal. Components of the apoptotic pathway have been shown to play a role in da neuron remodeling during metamorphosis (Kuo et al., 2006; Rumpf et al., 2011; Williams and Truman, 2005b; Williams et al., 2006). Moreover, Nos has previously been implicated in translational repression of the proapoptotic factor Hid in developing germ cells (Sato et al., 2007). Although Hid has not been implicated in pruning during metamorphosis, we hypothesized that the excessive retraction phenotype of nos and pum deficient neurons could be due to an inappropriate production of Hid and other apoptotic machinery in da neurons. Anti-Hid immunofluorescence detects little or no Hid protein in ddaC neurons in wild-type wandering larvae (Fig. 2A-C). In contrast, Hid levels are elevated in nos mutant, nosRNAi, and pum mutant neurons, with Hid often concentrated in one or two foci within the nucleus (Fig. 2D-I; Fig. S1). To determine whether this increase in Hid contributes to the loss of dendrites in nos deficient neurons, we tested the effect of lowering hid gene dosage on dendrite morphology in nosRNAi neurons. The loss of dendritic termini in nosRNAi da neurons is fully rescued when larvae are heterozygous for the H99 chromosomal deficiency that removes the hid locus (nosRNAi; H99/+; Fig. 2J-M). This effect is specific to nos deficient neurons, as dendrite number in ddAC neurons from H99/+ larvae is indistinguishable from wild-type (data not shown). Moreover, Hid levels are restored to wild-type levels in nosRNAi; H99/+ neurons, consistent with the restoration of branch termini (Fig. S1).

Fig. 2
Nos and Pum repress Hid expression. (A-I) Confocal z series projections of ddaC neurons from wild-type (WT; A-C), nos mutant (nos; D-F) and pum mutant (pum; G-I) larvae stained with anti-GFP antibody to detect the mCD8-GFP marker (green) ...

Upon dendrite severing, caspases are activated locally within the severed dendrites to promote their removal (Williams et al., 2006). We do not, however, detect activated caspase within nos or pum deficient ddaC neuron dendrites using immunofluorescence (data not shown). Consistent with this result, nos and pum mutant neurons do not exhibit the dendrite severing that is characteristic of pruning during metamorphosis. While we cannot definitively rule out a role for caspases in contributing to the nos mutant phenotype, our results suggest that hid is a target of repression by Nos and Pum in da neurons and that accumulation of Hid in nos and pum mutant neurons results in branch retraction by a caspase-independent mechanism.

In contrast to the long and evenly spaced dendritic branches of control ddaC neurons, nos deficient neurons have shorter terminal dendritic branches that tend to cluster together (Fig. 2N), suggesting that nos promotes branch length and regulates dendritic spacing. Lowering hid gene dosage does not rescue the defect in terminal branch length or branch spacing, however, and terminal branches are significantly shorter in nosRNAi/H99 neurons as compared to both control and nosRNAi neurons (Fig. 2J′,L′,N). Thus, while hid appears to act downstream of Nos to regulate dendrite maintenance and retraction, regulation of branch length and spacing is likely mediated by other Nos targets or cofactors.

Nos and Pum regulate Cut expression in Class IV da neurons

The defects associated with terminal dendrite length in nos deficient neurons occur independently of apoptotic machinery suggesting that additional Nos targets promote branch length. The distinct dendritic patterns and morphologies exhibited by different classes of da neurons are controlled in part by the level of the transcription factor Cut (Grueber et al., 2003, Li et al., 2004, Jinushi-Nakao et al., 2007). Class IV da neurons express intermediate levels of Cut protein during both embryonic and larval development (Grueber et al., 2003) suggesting that Cut functions not only during the initial differentiation of these neurons, but also during their subsequent development to specify their characteristic dendritic patterns. Similarly to nos and pum mutants, cut mutant class IV da neurons show reduced higher order dendritic branching and reduced branch length. Moreover, overexpression of nos and pum in class IV da neurons results in a severe loss of dendritic branching and a dramatic reduction in the length of individual branches (Ye et al., 2004); a similar defect is produced by cut overexpression (Grueber et al., 2003; Jinushi-Nakao et al., 2007). The similarity in these overexpression phenotypes suggest that Nos and Pum might regulate dendritic branching and promote branch length through an effect on cut expression. Anti-Cut immunofluorescence shows that overexpressing nos or pum using UAS-nos-tub3′UTR and UAS-pum transgenes, respectively, driven by GAL4477 results in an increase in Cut protein levels within ddaC neurons (Fig. 3A-C). We investigated whether the elevated Cut levels contribute to the loss of dendrites in neurons overexpressing nos or pum by genetically removing one copy of cut with the strong hypomorphic cut145 allele (Grueber et al., 2003). Lowering Cut levels in this way has no effect on its own but partially suppresses the loss of dendritic branches and branch length defects associated with nos and pum overexpression (Fig. 4A-E and data not shown), suggesting that cut functions downstream of nos and pum in dendrite development.

Fig. 3
Nos and Pum differentially influence Cut expression. Anti-Cut (magenta) and anti-GFP (green) immunostaining of ddaC neurons marked with mCD8-GFP. (A, A′) Wild-type (WT) ddaC neurons express intermediate levels of Cut. Cut levels are increased ...
Fig. 4
cut interacts genetically with nos and pum. (A-D) Larval ddaC neurons visualized using GAL4477 to express mCD8-GFP. GAL4477-mediated overexpression of nos (nosOE; A) or pum (pumOE; C) results in loss of higher order branches. Reduction of cut levels by ...

We next investigated whether Cut levels are affected in neurons deficient for nos or pum. Conversely to nos overexpression, nos mutant and nosRNAi neurons have decreased levels of Cut protein compared to wild-type neurons (Fig. 3D,E). In contrast, pum mutant neurons express Cut protein at higher levels than wild-type neurons (Fig. 3F). Thus, while the phenotypes due to overexpression of nos and pum may result in part from inappropriate upregulation of Cut expression, the different effects on Cut protein levels in nos versus pum mutant neurons suggest that Nos and Pum regulate distinct targets that affect Cut level differentially within ddaC neurons. GAL4/UAS-mediated overexpression of Cut produces a more severe phenotype than is observed in pum mutant neurons (Grueber et al., 2003; Jinushi-Nakao et al., 2007). However, modulation of Cut activity by other factors is required for generating the distinct dendritic patterns of the different classes of da neurons (Jinushi-Nakao et al., 2007) and these factors may be missing from pum mutant neurons.

brat is required for dendritic development of ddaC neurons

To survey the role of RNA-binding proteins in neuronal development, we performed an RNAi screen in which we systematically knocked down the expression of the majority of the annotated Drosophila RNA binding proteins specifically in class IV da neurons (E.C.O. and E.R.G. unpublished). The screen, in which UAS-RNAi transgenes were expressed in class IV da neurons using GAL4477, will be described elsewhere. Among the positive candidates from this screen, we identified brat. Knockdown of brat using either of two different UAS-bratRNAi transgenes results in a loss of higher order dendritic branches and a reduction in the overall field of coverage in ddaC neurons as compared to wild-type neurons (Fig. 5A,B and data not shown). Loss of higher order branching was also observed using another class IV da neuron-specific driver, ppk-GAL4 (data not shown).

Fig. 5
brat plays a cell autonomous role in da neuron dendrite morphogenesis. (A, B) Larval ddaC neurons expressing UAS-mCD8:GFP alone (A) or together with UAS-brat RNAi (B) using GAL4477. (C, D) Wild-type control (WT; C) and brat11 (D) ddaC MARCM clones with ...

A requirement for brat in dendrite morphogenesis was confirmed using two brat mutant alleles. Viable larvae were obtained for transheterozygous combinations of the brat1 and brat11 alleles and for the brat1 allele in trans to a deficiency spanning the brat locus. Similarly to brat RNAi, these mutant combinations result in a loss of higher order dendritic branching in ddaC neurons (Fig. 5F-H). Moreover, brat11 mutant clones generated by the Mosaic Analysis with a Repressible Cell Marker (MARCM) method (Lee and Luo, 2001) show reduced dendritic complexity and field coverage as compared to control ddaC mitotic clones (Fig. 5C,D,E), confirming that brat is required cell autonomously in class IV da neurons for dendrite development. Consistent with these results, anti-Brat immunofluorescence detects cytoplasmic Brat protein in the ddaC neuron cell body (Fig. 5I-K).

In addition to class IV da neurons, class III da neurons require nos and pum for dendrite development. In contrast, nos and pum are dispensable in class I and class II da neurons (Ye et al., 2004). Analysis of brat11 MARCM clones shows no effect on the development of class I, II and III da neurons (Fig. S2). Thus, the requirement for brat appears to be limited to class IV da neurons, although we cannot rule out the possibility that residual function of the hypomorphic brat11 allele suffices in the other classes.

brat functions with nos and pum to regulate dendrite development

The requirement for brat in class IV da neurons suggests that Brat may function in concert with Nos and Pum to mediate translational repression during dendrite morphogenesis. We therefore tested whether brat interacts genetically with nos and pum in da neurons. Overexpression of brat using a UAS-brat transgene driven by GAL4477 results in reduced dendritic complexity and field coverage (Fig. 6A) and similar results were obtained with ppk-GAL4 (Fig. S4E,F,H). Reducing nos function using RNAi or by mutation of one copy of nos suppresses the dendritic defects caused by brat overexpression (Fig. 6B,C,K). Similarly, reducing pum function suppresses dendritic defects associated with brat overexpression (Fig. 6D,E,K).

Fig. 6
nos, pum and brat genetically interact to promote dendrite morphogenesis. (A-J) Larval ddaC neurons overexpressing brat (bratOE), nos (nosOE) or pum (pumOE). (A-E) GAL4477 was used to express both UAS-mCD8:GFP and UAS-brat. Neurons overexpressing brat ...

The suppression of the brat overexpression phenotype by reducing nos and pum function suggests that brat functions either upstream of or together with nos and pum to regulate dendrite morphogenesis. In order to distinguish between these two possibilities, we took advantage of nos and pum overexpression phenotypes to perform epistasis experiments. nos and pum overexpression driven by ppk-GAL4 results in a dramatic loss of dendritic branching (Fig. 6F,I respectively). Removing one copy of brat has no effect on its own, but partially suppresses the nos overexpression phenotype and more substantially suppresses the pum overexpression phenotype (Fig. 6G,J,K and data not shown). Such partial suppression of nos overexpression is also observed when pum function is reduced (Fig. 6H). We suspect that nos overexpression can cause some degree of toxicity, as is observed in other tissues (Clark et al., 2002). Together, these results provide evidence that brat, pum and nos function together to regulate dendritic development.

Hid but not Cut expression is affected in brat deficient neurons

Results presented above suggest that Nos and Pum repress translation of hid in da neurons. To determine whether Brat also contributes to regulation of hid, we assayed Hid protein levels in brat mutant neurons. Anti-Hid immunofluorescence shows elevated Hid protein levels in neurons from brat11/Df and brat11/brat1 larvae as compared to wild-type neurons (Fig. S3), consistent with repression of hid by a complex involving Nos, Pum, and Brat.

We investigated whether Brat, like Nos and Pum, might also regulate cut expression in da neurons. Anti-Cut immunofluorescence shows that brat overexpression results in elevated levels of Cut protein in ddaC neurons, similar to neurons overexpressing either nos or pum. However, Cut expression is unaffected in brat mutant neurons (Fig. S3). Additionally, reducing cut function does not suppress the loss of higher order branches in neurons overexpressing brat (Fig. S4A,B,D). Taken together, our data suggest that brat is not involved in regulating Cut expression.

d4EHP interacts with brat during dendrite morphogenesis

In the early Drosophila embryo, the Nos/Pum/Brat complex represses translation of hb mRNA by a cap-dependent mechanism. d4EHP, an eIF4E-like cap binding protein, facilitates this repression by simultaneously interacting with Brat and the 7-methyl guanosine cap structure at the 5′ end of hb mRNA (Cho et al., 2006). In our RNAi screen, d4EHP RNAi produced a reproducible but weak defect in the dendrite arborization pattern, where termini are shortened and unevenly distributed creating dendrite free regions within the dendritic tree (data not shown). In larvae mutant for the viable d4EHPCP53 mutant allele, the number of dendritic termini is not affected in ddaC neurons, but they display defects in dendrite patterning whereby branches are shortened and unevenly distributed, similar to d4EHPRNAi neurons (Fig. S5). Due to the hypomorphic nature of the d4EHPCP53 allele and the potentially incomplete knockdown of d4EHP via RNAi, it remains possible that a null mutant allele would produce more severe dendritic defects.

We next investigated whether d4EHP interacts genetically with brat to regulate dendrite morphogenesis in ddaC neurons. Reducing d4EHP function partially rescues the dendritic branching defect caused by brat overexpression (Fig. S4A,C,D). Moreover, overexpression of a mutant form of brat that is unable to bind d4EHP (Harris et al., 2011) results in a weaker dendrite morphogenesis defect than overexpression of wild-type brat (Fig. S4F,G,H). Taken together, the overexpression and epistasis analyses show that brat genetically interacts with d4EHP in da neurons to regulate dendrite development.

Brat functions presynaptically at the NMJ to regulate bouton morphogenesis

In contrast to their collaborative role in da neurons, Nos and Pum function in opposition to one another to regulate both bouton morphogenesis and the electrophysiological properties of the NMJ (Menon et al., 2009). Anti-Brat immunofluorescence detects Brat in the larval NMJ, suggesting a role for Brat in NMJ development. To begin to distinguish how Brat might function at the NMJ, we examined Brat localization relative to the presynaptic marker, Synaptotagmin-GFP (Syt-GFP) and the postsynaptic marker Discs large-YFP (Dlg-YFP). Immunofluorescence detection of Brat together with each marker shows that Brat colocalizes presynaptically with Syt-GFP, but not postsynaptically with Dlg-YFP (Fig. 7A-F).

Fig. 7
brat function is required for bouton morphogenesis at the larval NMJ. Confocal z series projections of larval NMJs from muscle 6/7. (A-F) Immunofluorescence detection of Brat (magenta) together with (A) the presynaptic marker Syt-GFP (green) or (D) the ...

To determine whether the presence of Brat in the NMJ reflects its function there, we examined brat11/brat1 and brat11/Df third instar larvae immunostained for either endogenous Syt or Dlg to visualize boutons. Boutons in brat mutant NMJs are often misshapen, fused or enlarged as compared to wild-type NMJs (Fig. 7J-L). Quantification of bouton number, most easily performed using anti-Syt immunofluorescence, shows a statistically significant reduction in bouton number in brat mutant relative to wild-type NMJs (Fig. 7N). This defect is similar to the defect observed in pum mutant NMJs (Menon et al., 2004).

We targeted brat RNAi to the pre- or post-synaptic compartments to ascertain where brat function is required. Presynaptic expression of brat RNAi using OK6-GAL4 produces a pum-like phenotype, with a reduction in bouton number, as well as misshapen, enlarged or fused boutons as compared to wild-type NMJs. Similar results were obtained with two different UAS-brat RNAi lines (Fig. 7G-I,M). In contrast, brat RNAi expressed postsynaptically, using either the Mef2-GAL4 or Mhc82-GAL4 muscle drivers, does not produce overt bouton defects (data not shown). Both Mef2-GAL4 and Mhc82-GAL4 effectively drive the expression of a UAS-mCherry reporter indicating that they are indeed functional (data not shown). These results indicate that brat functions presynaptically to regulate bouton morphogenesis in the NMJ, consistent with the presynaptic localization of Brat protein. We conclude that brat plays a fundamental role in the development of synaptic boutons in the larval neuromuscular system.

The brat mutant NMJ phenotype resembles the phenotype of pum rather than nos mutant NMJs. We were unable, however, to test genetically whether brat functions with pum in bouton morphogenesis. Overexpression of brat either pre- or postsynaptically does not cause bouton morphogenesis defects (data not shown). While presynaptic pum overexpression does produce a bouton defect (Menon et al., 2004), we were unable to generate animals with presynaptic pum overexpression that were also heterozygous for brat mutations. Nonetheless, our results suggest that brat functions presynaptically to regulate bouton morphogenesis and the similarity of the brat and pum mutant bouton phenotypes suggests that Brat and Pum function together, in opposition to Nos, to regulate bouton development in the larval NMJ.


Post-transcriptional mechanisms of gene regulation such as translational control play a fundamental role in the development and function of the nervous system (Eberwine et al., 2001; Mikl et al., 2010; Schuman et al., 2006; Job et al., 2001; Martin, 2004). Genetic studies have identified roles for the translational repressors Nos and Pum in sensory neuron and NMJ morphogenesis, NMJ function, and motor neuron excitability, and Pum has been implicated in long-term memory (Baines 2005). Understanding the selectivity of these regulators for different mRNA targets is essential to identify the cellular processes they regulate for neuronal morphogenesis and neural function. Here, we show that different combinations of Nos, Pum, and the co-factor Brat confer cell type-specific regulation during morphogenesis of Drosophila da sensory neurons and the NMJ.

In Drosophila class IV da neurons, dendritic arbors grow rapidly during the first larval instar to establish nonredundant territories that cover the larval body wall. During the second and third larval instars, da neuron dendrites add and lengthen higher order branches to maintain body wall coverage as the larva undergoes dramatic growth. Results from our live imaging analysis place the requirement for Nos and Pum during the third larval instar, indicating that Nos and Pum are not involved in the establishment of dendritic territories but rather in maintaining the density of terminal branches during late larval growth by promoting branch extension and preventing branch retraction. We cannot, however, rule out the possibility that branch stabilization depends on Nos and Pum activity earlier during larval development. We also provide evidence that this maintenance function of Nos and Pum depends on their regulation of the proapoptotic protein Hid. Nos has previously been proposed to repress hid mRNA translation in developing germ cells to suppress apoptosis, although requirements for Pum and Brat were not tested (Sato et al., 2007). Together, our data showing that Hid is elevated in nos and pum mutant da neurons and that both the upregulation of Hid and the loss of terminal branches in nos mutants are suppressed by reduction of hid gene dosage suggest that repression of hid mRNA translation by Nos and Pum is also crucial for dendrite morphogenesis. Biochemical analysis will be required to test this model directly.

In cultured Drosophila cells, Hid localizes to mitochondria and this localization is required for full caspase activation (Haining et al., 1999; Abdelwahid et al. 2007). By contrast, Hid protein is detected in the nucleus in nos and pum mutants. A similar nuclear accumulation has been proposed to sequester Hid in larval malphigian tubules and prevent apoptosis of this tissue during metamorphosis (Shukla and Tapadia, 2011). The nuclear accumulation of Hid may indeed explain why upregulation of Hid in nos and pum da mutants does not cause cell death. Nuclear Hid sequestration in nos and pum mutant neurons is also consistent with the apparent absence of activated caspase. How Hid causes dendrite loss in nos and pum mutant neurons remains to be determined but could involve activation of a pathway similar to injury induced dendrite degeneration, which resembles pruning but is caspase-independent (Tao and Rolls, 2011).

Nos and Pum were initially identified because of their role in translational repression of hb mRNA in the posterior region of the early embryo. There, the two proteins form an obligate repression complex, with Pum conferring the RNA-binding specificity and Nos, which is synthesized only at the posterior pole of the embryo, providing the spatial specificity (Thompson et al., 2007). More recent studies have shown that Nos and Pum are not obligate partners, however. In the ovary, Pum functions together with Nos in germline stem cells to promote their self-renewal, while Pum acts independently of Nos in progeny cystoblasts to promote their differentiation (Harris et al., 2011). In the NMJ, Pum and Nos work in opposition to one another to regulate both morphological and electrophysiological characteristics of synaptic boutons (Menon et al., 2009). While Hid levels are similarly elevated in nos and pum mutant da neurons, the differential effects on cut expression observed in the two mutants suggest that in addition to working together, Nos and Pum participate in separate complexes that target different mRNAs even within the same cell type. Presumably, additional factors that associate selectively with Nos or Pum drive the formation of distinct complexes with different binding specificities. Pum represses eIF4E in the post-synaptic NMJ independently of Nos (Menon et al., 2004), suggesting that some of Pum’s effects in da neurons could be through more global effects on translation.

A third cofactor, Brat, is required for Nos/Pum-dependent repression of hb mRNA in the early embryo and paralytic mRNA in motorneurons (Sonoda and Wharton, 2001; Muraro et al., 2008). However, Brat is not required for Nos/Pum-mediated repression of cyclin B mRNA in primordial germ cells or for Nos/Pum function in germline stem-cell maintenance (Sonoda and Wharton, 2001; Harris et al., 2011). Structural and molecular analyses have shown that Brat is recruited to the Nos/Pum/NRE ternary complexes through an interaction between its conserved NHL (NCL-1, HT2A, and LIN-41) domain and Pum (Edwards et al., 2003). The Brat NHL domain also mediates interaction of Brat with the eIF4E-binding protein d4EHP and mutations in Brat that abrogate this interaction partially disrupt translational repression of hb, suggesting a mechanism by which the Pum/Nos/Brat/NRE complex could repress cap-dependent initiation (Cho et al., 2006). Our results indicate that Brat also collaborates with Nos and Pum to regulate dendrite morphogenesis by a mechanism involving d4EHP interaction and that this requirement is cell type-specific. While genetic analysis suggests that Brat is required for Nos/Pum-mediated regulation of dendrite complexity and Hid expression in class IV da neurons, it is dispensible for Nos and Pum function in class III da neurons. A similar cell type-specific requirement for Brat function in Nos/Pum-mediated repression within the CNS has been proposed based on the ability of brat mutants to counteract repression of paralytic mRNA due to Pum overexpression (Muraro et al., 2008). Since Brat is expressed throughout the dorsal cluster of larval sensory neurons (data not shown) and CNS, it is unclear whether the recruitment of Brat to the complex occurs only in certain cell types or whether its function in the complex is target dependent. In contrast to nos and pum mutants, however, brat mutants have no effect on cut expression, suggesting that Brat’s role in translational regulation is in fact limited to a subset of Nos/Pum-dependent processes.

Our findings that Brat functions presynaptically in bouton formation and that brat and pum mutant NMJs exhibit similar defects in bouton formation suggest that Brat is selectively recruited by Pum, but not by Nos, to regulate distinct target mRNAs in bouton development. Similarly, Brat functions selectively with Pum in ovarian cystoblasts to promote differentiation (Harris et al., 2011), suggesting that a Pum/Nos/NRE ternary complex is not essential for recruitment of Brat. Pum and many of its homologs in other organisms, members of the large Puf (Pum/FBF) protein family, typically recognize sequences that contain a core UGUA motif, although features beyond the core element also influence target mRNA recognition (Bernstein et al., 2005). We have shown that Pum can also recognize a UGUG motif that is found in binding sites for the C. elegans Puf protein FBF (Menon et al, 2009). Thus, it is possible that the interaction of Pum with different binding sites dictates the assembly of the particular repression complex. Interactors like Brat might add an additional layer of regulation by altering the specificity or affinity of Pum for particular targets, thereby generating diverse cellular and morphological outputs within a particular cell type.


  • Nanos and Pumilio promote dendrite outgrowth and stabilize existing dendrite branches
  • Nanos and Pumilio regulate Head Involution Defective and Cut expression in da neurons
  • Brain Tumor interacts genetically with Nanos and Pumilio for dendrite morphogenesis

Supplementary Material


Figure S1. Upregulation of Hid in nosRNAi neurons is suppressed by the H99 deficiency. (A-I) Confocal Z series projections of ddaC neurons from wild-type larvae (WT; A-C), nosRNAi larvae (D-F), and nosRNAi larvae heterozygous for the H99 deficiency (nosRNAi; H99/+; G-I). UAS-mCD8:GFP and UAS-nosRNAi transgenes were expressed in class IV da neurons using GAL4477. Larvae were immunostained with anti-GFP antibody to detect the mCD8-GFP marker (green) and anti-Hid antibody (magenta). Enlargements of the region of the cell body are shown. Nuclei from representative ddaC neurons for each genotype are indicated with arrows. The intensity of the green channel was reduced in the merged images (C, F, I) to visualize Hid expression. Scale bar represents 50 μm. (J) Quantification of fluorescence intensity showed that Hid levels are significantly upregulated in nosRNAi neurons compared to WT samples, but expression is restored to control levels in nosRNAi; H99/+ neurons. n=15 for each genotype. **p<0.01, ***P<0.001, analyzed with Student’s t-test.


Fig. S2. brat function is dispensible for other classes of da neurons. (A-F) MARCM clones of class I (A, B), class II (C, D), and class III (E,F) da neurons. Wild-type control clones (A,C,E) and brat11 clones (B, D, F) are indistinguishable.


Fig. S3. Brat regulates Hid but not Cut expression. (A, B) Anti-Hid (magenta) and anti-GFP (green) immunostaining of ddaC neurons marked with mCD8-GFP in wild-type (WT; A, A′) and brat11/Df (B, B′) larvae. Arrows indicate ddaC nuclei. Quantification of fluorescence intensity (see Materials and Methods) shows that relative to wild-type neurons, Hid is upregulated 2 fold in brat mutant ddaC neurons (p=0.002; n=18). (C, D) Anti-Cut (magenta) and anti-GFP (green) immunostaining of ddaC neurons marked with mCD8-GFP in wild-type (C, C′) and brat1/brat11 larvae. Nuclei of ddaC neurons are indicated with arrows and class III neurons, which express high levels of Cut, are indicated with an asterisk. Cut levels are not statistically different between wild-type and brat mutant ddaC neurons.


Fig. S4. cut and d4EHP genetically interact with brat in da neuron development. (A-C) Larval ddaC neurons expressing UAS-mCD8:GFP and UAS-brat under control of GAL4477. Reducing cut dosage rescues dendrite coverage defects but not the defects in termini number associated with brat overexpression. Reducing d4EHP dosage rescues dendrite coverage defects and partially rescues defects in termini number associated with brat overexpression. (D) Quantification of the total number of terminal branches: bratOE (n=18), ctc145/+; bratOE/+ (n=20), d4EHPCP53/bratOE (n=27). (G) Overexpressing a mutant form of brat (bratRD) that does not interact with d4EHP, using ppk-Gal4 causes a decrease in terminal branch number compared to (E) control neurons. Overexpressing wild type brat using ppk-Gal4 (F) causes a decrease in terminal branch number compared to (E) control neurons and neurons expressing bratRD. (H) Quantification of the total number of terminal branches: control (n=17), bratWT (n=20), bratRD (n=26).


Fig. S5. Dendrite patterning defects in d4EHP mutant neurons. ddaC neurons labeled with mCD8-GFP in (A) wild-type (WT) larva and (B) d4EHPCP53 homozygous mutant larva.


We thank H. Ashe, A. Brand, N. Brown, I. Davis, W. Grueber, Y.N. Jan, K. Menon, D. St Johnston, R. Wharton, K. Zinn, the Bloomington Drosophila stock center, the Vienna Drosophila RNAi Center, and TRiP at Harvard Medical School (NIH/NIGMS R01-GM084947) for fly stocks and P. Lasko, H. Steller, and R. Wharton for antibodies. This work was supported by the National Institutes of Health (F32 HD056779 to E.C.O and R01 GM061107 to E.R.G).


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Abbott MK, Lengyel JA. Embryonic head involution and rotation of male terminalia require the Drosophila locus head involution defective. Genetics. 1991;129:783–789. [PubMed]
  • Abdelwahid E, Yokokura T, Krieser RJ, Balasundaram S, Fowle WH, White K. Mitochondrial disruption in Drosophila apoptosis. Dev Cell. 2007;12:793–806. [PubMed]
  • Aberle H, Haghighi AP, Fetter RD, McCabe BD, Magalhães TR, Goodman CS. Wishful thinking encodes a BMP type II receptor that regulates synaptic growth in Drosophila. Neuron. 2002;33:545–558. [PubMed]
  • Baines RA. Neuronal homeostasis through translational control. Mol Neurobiol. 2005;32:113–121. [PubMed]
  • Bassell GJ, Warren ST. Fragile X Syndrome: loss of local mRNA regulation alters synaptic development and function. Neuron. 2008;60:201–214. [PMC free article] [PubMed]
  • Bernstein D, Hook B, Hajarnavis A, Opperman L, Wickens M. Binding specificity and mRNA targets of a C. elegans PUF protein, FBF-1. RNA. 2005;11:447–458. [PubMed]
  • Brechbiel JL, Gavis ER. Spatial regulation of nanos is required for its function in dendrite morphogenesis. Curr Biol. 2008;18:745–750. [PMC free article] [PubMed]
  • Cho PF, Gamberi C, Cho-Park YA, Cho-Park IB, Lasko P, Sonenberg N. Cap-dependent translational inhibition establishes two opposing morphogen gradients in Drosophila embryos. Curr Biol. 2006;16:2035–2041. [PMC free article] [PubMed]
  • Clark IE, Dobi KC, Duchow HK, Vlasak AN, Gavis ER. A common translational control mechanism functions in axial patterning and neuroendocrine signaling in Drosophila. Development. 2002;129:3325–3334. [PubMed]
  • Curtis D, Treiber DK, Tao F, Zamore PD, Williamson JR, Lehmann R. A CCHC metal-binding domain in Nanos is essential for translational regulation. EMBO J. 1997;16:834–843. [PubMed]
  • Dietzl G, Chen D, Schnorrer F, Su KC, Barinova Y, Fellner M, Gasser B, Kinsey K, Oppel S, Scheiblauer S, Couto A, Marra V, Keleman K, Dickson BJ. A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature. 2007;448:151–156. [PubMed]
  • Eberwine J, Miyashiro K, Kacharmina JE, Job C. Local translation of classes of mRNAs that are targeted to neuronal dendrites. Proc Natl Acad Sci USA. 2001;98:7080–7085. [PubMed]
  • Edwards TA, Wilkinson BD, Wharton RP, Aggarwal AK. Model of the Brain Tumor-Pumilio translational repressor complex. Genes Dev. 2003;17:2508–2513. [PubMed]
  • Fallini C, Zhang H, Su Y, Silani V, Singer RH, Rossoll W, Bassell GJ. The survival of motor neuron (SMN) protein interacts with the mRNA-binding protein HuD and regulates localization of poly(A) mRNA in primary motor neuron axons. J Neurosci. 2011;31:3914–3925. [PMC free article] [PubMed]
  • Forbes A, Lehmann R. Nanos and Pumilio have critical roles in the development and function of Drosophila germline stem cells. Development. 1998;125:679–690. [PubMed]
  • Frank DJ, Edgar BA, Roth MB. The Drosophila melanogaster gene brain tumor negatively regulates cell growth and ribosomal RNA synthesis. Development. 2002;129:399–407. [PubMed]
  • Grueber WB, Jan LY, Jan YN. Different levels of the homeodomain protein Cut regulate distinct dendrite branching patterns of Drosophila multidendritic neurons. Cell. 2003;112:805–818. [PubMed]
  • Grueber WB, Ye B, Yang CH, Younger S, Borden K, Jan LY, Jan YN. Projections of Drosophila multidendritic neurons in the central nervous system: links with peripheral dendrite morphology. Development. 2007;134:55–64. [PubMed]
  • Haining WN, Carboy-Newcomb C, Wei CL, Steller H. The proapoptotic function of Drosophila Hid is conserved in mammalian cells. Proc Natl Acad Sci USA. 1999;96:4936–4941. [PubMed]
  • Harris RE, Pargett M, Sutcliffe C, Umulis D, Ashe HL. Brat promotes stem cell differentiation via control of a bistable switch that restricts BMP signaling. Dev Cell. 2011;20:72–83. [PMC free article] [PubMed]
  • Jinushi-Nakao S, Arvind R, Amikura R, Kinameri E, Liu AW, Moore AW. Knot/Collier and Cut control different aspects of dendrite cytoskeleton and synergize to define final arbor shape. Neuron. 2007;56:963–978. [PubMed]
  • Job C, Eberwine J. Localization and translation of mRNA in dendrites and axons. Nat Rev Neurosci. 2001;2:889–898. [PubMed]
  • Kuo CT, Zhu S, Younger S, Jan LY, Jan YN. Identification of E2/E3 ubiquitinating enzymes and caspase activity regulating Drosophila sensory neuron dendrite pruning. Neuron. 2006;51:283–290. [PubMed]
  • Lee A, Li W, Xu K, Bogert BA, Su K, Gao FB. Control of dendritic development by the Drosophila fragile X-related gene involves the small GTPase Rac1. Development. 2003;130:5543–5552. [PubMed]
  • Lee T, Luo L. Mosaic analysis with a repressible cell marker (MARCM) for Drosophila neural development. Trends Neurosci. 2001;24:251–254. [PubMed]
  • Li W, Wang F, Menut L, Gao FB. BTB/POZ-zinc finger protein Abrupt suppresses dendritic branching in a neuronal subtype-specific and dosage-dependent manner. Neuron. 2004;43:823–834. [PubMed]
  • Marek KW, Ng N, Fetter R, Smolik S, Goodman CS, Davis GW. A genetic analysis of synaptic development: pre- and postsynaptic dCBP control transmitter release at the Drosophila NMJ. Neuron. 2000;25:537–547. [PubMed]
  • Martin KC. Local protein synthesis during axon guidance and synaptic plasticity. Curr Opin Neurobiol. 2004;14:305–310. [PubMed]
  • Meijering E, Jacob M, Sarria JCF, Steiner P, Hirling H, Unser M. Design and Validation of a Tool for Neurite Tracing and Analysis in Fluorescence Microscopy Images. Cytometry Part A. 2004;58:167–176. [PubMed]
  • Menon KP, Andrews S, Murthy M, Gavis ER, Zinn K. The translational repressors Nanos and Pumilio have divergent effects on presynaptic terminal growth and postsynaptic glutamate receptor subunit composition. J Neurosci. 2009;29:5558–5572. [PMC free article] [PubMed]
  • Menon KP, Sanyal S, Habara Y, Sanchez R, Wharton RP, Ramaswami M, et al. The translational repressor Pumilio regulates presynaptic morphology and controls postsynaptic accumulation of translation factor eIF-4E. Neuron. 2004;44:663–676. [PubMed]
  • Mikl M, Vendra G, Doyle M, Kiebler MA. RNA localization in neurite morphogenesis and synaptic regulation: current evidence and novel approaches. J Comp Physiol. 2010;196:321–334. [PMC free article] [PubMed]
  • Muraro NI, Weston AJ, Gerber AP, Luschnig S, Moffat KG, Baines RA. Pumilio binds para mRNA and requires Nanos and Brat to regulate sodium current in Drosophila motoneurons. J Neurosci. 2008;28:2099–2109. [PMC free article] [PubMed]
  • Parisi M, Lin H. Translational repression: a duet of Nanos and Pumilio. Curr Biol. 2000;10:R81–R83. [PubMed]
  • Parrish JZ, Emoto K, Kim MD, Jan YN. Mechanisms that regulate establishment, maintenance, and remodeling of dendritic fields. Annu Rev Neurosci. 2007;30:399–423. [PubMed]
  • Parrish JZ, Xu P, Kim CC, Jan LY, Jan YN. The microRNA bantam functions in epithelial cells to regulate scaling growth of dendrite arbors in Drosophila sensory neurons. Neuron. 2009;63:788–802. [PMC free article] [PubMed]
  • Ranganayakulu G, Schulz RA, Olson EN. Wingless signaling induces nautilus expression in the ventral mesoderm of the Drosophila embryo. Dev Biol. 1996;176:143–148. [PubMed]
  • Rees JS, Lowe N, Armean IM, Roote J, Johnson G, Drummond E, Spriggs H, Ryder E, Russell S, St Johnston D, Lilley KS. In vivo analysis of proteomes and interactomes using parallel affinity capture (iPAC) coupled to mass spectrometry. Mol Cell Proteomics. 2011;10:M110.002386. [PMC free article] [PubMed]
  • Rumpf S, Lee SB, Jan LY, Jan YN. Neuronal remodeling and apoptosis require VCP-dependent degradation of the apoptosis inhibitor DIAP1. Development. 2011;138:1153–1160. [PubMed]
  • Sato K, Hayashi Y, Ninomiya Y, Shigenobu S, Arita K, Mukai M, Kobayashi S. Maternal Nanos represses hid/skl-dependent apoptosis to maintain the germ line in Drosophila embryos. Proc Natl Acad Sci USA. 2007;104:7455–7460. [PubMed]
  • Schuman EM, Dynes JL, Steward O. Synaptic regulation of translation of dendritic mRNAs. J Neurosci. 2006;26:7143–7146. [PubMed]
  • Shukla A, Tapadia MG. Differential localization and processing of apoptotic proteins in Malpighian tubules of Drosophila during metamorphosis. Eur J Cell Biol. 2011;90:72–80. [PubMed]
  • Sonoda J, Wharton RP. Drosophila Brain Tumor is a translational repressor. Genes Dev. 2001;15:762–773. [PubMed]
  • Tao J, Rolls MM. Dendrites have a rapid program of injury-induced degeneration that is molecularly distinct from developmental pruning. J Neurosci. 2011;31:5398–5405. [PMC free article] [PubMed]
  • Thompson B, Wickens M, Kimble J. Translational control in development. In: Mathews MB, Sonenberg N, Hershey JWB, editors. Translational Control in Biology and Medicine. Cold Spring Harbor Laboratory Press; NY: 2007. pp. 507–544.
  • Williams DW, Truman JW. Cellular mechanisms of dendrite pruning in Drosophila: insights from in vivo time-lapse of remodeling dendritic arborizing sensory neurons. Development. 2005a;132:3631–3642. [PubMed]
  • Williams DW, Truman JW. Remodeling dendrites during insect metamorphosis. J Neurobiol. 2005b;64:24–33. [PubMed]
  • Williams DW, Kondo S, Krzyzanowska A, Hiromi Y, Truman JW. Local caspase activity directs engulfment of dendrites during pruning. Nat Neurosci. 2006;9:1234–1236. [PubMed]
  • Wu CW, Zeng F, Eberwine J. mRNA transport to and translation in neuronal dendrites. Anal Bioanalytical Chemistry. 2007;387:59–62. [PubMed]
  • Ye B, Petritsch C, Clark IE, Gavis ER, Jan LY, Jan YN. Nanos and Pumilio are essential for dendrite morphogenesis in Drosophila peripheral neurons. Curr Biol. 2004;14:314–321. [PubMed]