The microRNA bantam functions in epithelial cells to regulate scaling growth of dendrite arbors in drosophila sensory neurons

Abstract

In addition to establishing dendritic coverage of the receptive field, neurons need to adjust their dendritic arbors to match changes of the receptive field. Here, we show that dendrite arborization (da) sensory neurons establish dendritic coverage of the body wall early in Drosophila larval development and then grow in precise proportion to their substrate, the underlying body wall epithelium, as the larva more than triples in length. This phenomenon, referred to as scaling growth of dendrites, requires the function of the microRNA (miRNA) bantam (ban) in the epithelial cells rather than the da neurons themselves. We further show that ban in epithelial cells dampens Akt kinase activity in adjacent neurons to influence dendrite growth. This signaling between epithelial cells and neurons receiving sensory input from the body wall synchronizes their growth to ensure proper dendritic coverage of the receptive field.

Figure 1

Dendrites of da neurons grow rapidly to establish receptive field coverage then grow in proportion to larval growth to maintain this coverage. (A) Live images of class IV neurons (ppk>mCD8-RFP) and epithelial cells (Arm-GFP) at 16, 48, and 96 hr AEL. The hatched red line denotes the dorsal midline in (A) and in subsequent images. Dorsal is up, anterior is left and scale bars are 50 μm. (B) Quantification of the coverage index of an individual ddaC class IV neuron over a developmental time course. n > 10 for each time point. (C) Live images of Drosophila larvae over the course of larval development. (D) Quantification of the area covered by an individual ddaC class IV neuron over a developmental time course. n > 5 for each time point. (E) Time course showing the fold-increase in area of an individual ddaC dendritic field (red), an individual epithelial cell (green), and the ddaC cell body (blue) relative to the area of each at 16 hr AEL, shortly after dendrite outgrowth begins. n > 10 for each time point. (F) Time course of dendrite coverage for different classes of da neurons. Class IV neurons (green) were visualized with ppk-eGFP, class III neurons (red) were visualized with ss-Gal4, UAS-mCD8-GFP (M. Kim, personal communication) and class I neurons were visualized with 2-21-Gal4, UAS-mCD8-GFP (Grueber et al., 2003a). n > 10 for each time point.

Figure 2

Screen for mutants that affect scaling growth of dendrites. (A) Growth-defective mutants at the time point where dendrite growth was assayed. ppk>mCD8-GFP was used to visualize dendrites for 3^rd chromosome and ppk-EGFP was used to visualize dendrites for 2^nd chromosome mutants. (B) Dendrites scale to accommodate diverse sizes of receptive fields, as illustrated by class IV dendrites of pdp, G-5α, twf and wt larvae at 96 hr AEL. (C) Dendrites scale properly in mutants defective in developmental timing, such as b6-22, which persist as 2^nd instar larvae (S. Younger, personal communication), or persistent 3^rd instar gt or l(2)gl larvae. For (B) and (C) scale bars are 50 μm. (D-F) Effects of growth-defective mutants on dendrite development.(D) Segment width (average length of a dorsal hemisegment along the AP axis), (E) Hemisegment area (average area of the dorsal region of a hemisegment), and (F) Coverage index of mutants at the indicated time points. Note that although the size/shape of the receptive field is changed in most cases, only ban, and to a lesser degree InR and Akt, showed a substantial increase in coverage index. Complete gene names are provided in Table S1. Measurements were taken from dendrites of at least 4 neurons for each mutant.

Figure 3

ban is required for scaling growth of dendrites in a subset of da neurons. (A) Growth profile of control and ban mutant larvae. (B) Dendrite morphology of wt and ban mutant larvae at 96 hr AEL visualized with ppk-EGFP. The interface of two ddaC neurons at the dorsal midline is shown in (B’) and dendrites are depicted as camera lucida to help distinguish dendrites of adjacent neurons in (B”). Scale bars are 50 μm. (C) Dendrite morphology of wt and ban mutant larvae at 48 hr AEL visualized with ppk-EGFP. (D-E) Time course monitoring the dendrite coverage index (D) and midline crossing (E) of ddaC dendrites in wt or ban mutant larvae. n > 5 for all time points. (F-G) Coverage index for (F) ddaF class III neurons and (G) ddaE class I neurons at 96 hr AEL. n = 10. (H) Quantification of terminal dendrite dynamics measured over a 24 hr time lapse (72-96 hr AEL) in wt and ban mutant larvae. n > 1000 terminal dendrites (from 10 neurons) for each genotype. Here and in subsequent figures, error bars represent SEM, * denotes p < 0.05 and ** denotes p < 0.001.

Figure 4

ban regulates growth-inhibitory signals to ensure dendrite scaling. (A-D) Progressive restriction of invasion capacity of class IV dendrites. Class IV neurons were ablated with a focused laser beam at the time indicated and larvae were imaged 48 hr post-ablation. Hatched red lines demarcate the dorsal midline and segment boundaries. (D) Quantification of the portion of unoccupied dendritic territory covered by invading dendrites (invasion index). n > 10 for each ablation/imaging paradigm. (E-F) Dendrite coverage is maintained by dendrite scaling, even when dendrites establish aberrant body wall coverage. Class IV neurons were ablated at 24 hr AEL and larvae were imaged at 24 hr intervals beginning at 48 hr AEL. Images of the same territory at 48 hr and 96 hr AEL are shown. Invading dendrites from neighboring ddaC neurons are depicted in blue in the traces below each image. Hatched red lines demarcate the dorsal midline (dorsal is up) and segment boundaries. Dendrites invading the ventral portion of the field at 96 hr AEL are not traced because they are out of the field of view at 48 hr AEL. (F) The invasion index for each individual neuron monitored in this ablation paradigm is plotted at 48 and 96 hr AEL with black lines connecting the values for a given neuron. Mean values for 10 neurons are shown in red. (G-I) Dendrites retain exuberant invasion capacity in ban mutants. Invasion activity in ban mutants (G) at 96 hr AEL in which a class IV neuron was ablated at 48 hr AEL or (H) at 120 hr AEL in a ban mutant in which a class IV neuron was ablated at 72 hr AEL. (I) Quantification of invasion activity. n > 10 for each genotype. Scale bars are 50 μm for all images.

Figure 5

ban activity is broadly distributed in late larval stages but is dispensible in neurons for dendrite scaling. (A) A control miRNA sensor is ubiquitously expressed throughout larval development. Anti-GFP immunostaining of a control miRNA sensor 3^rd instar fillet reveals expression in muscle (arrow), epithelia (open arrowhead), and PNS neurons (bracket), which are labeled by anti-HRP immunoreactivity in magenta (A’). Scale bars are 50 μm for (A-E). (B-F) ban sensor expression in larvae. ban sensor GFP expression is not detectably expressed in 3^rd instar larvae (B). In (B’) ban sensor expression (green) and HRP (magenta) are shown. (C-D) Live imaging of the ban sensor in larvae using identical microscope settings at 48 hr AEL (E) and 72 hr AEL (F) shows a marked down-regulation of GFP expression over this interval. (E) ban sensor expression remains high in ban mutant larvae, even at 96 hr AEL, demonstrating that the down-regulation of the sensor seen in (D) and (E) depends on ban function. (F) Live imaging of the ban sensor in ban^△1 ddaC MARCM clones to monitor ban activity in clones. The ban sensor (green) and the CD2-Cherry used to positively label the clone (magenta) are shown. (G-I) ban is dispensable in neurons for dendrite scaling. Control (G) or ban^△1 (H) ddaC MARCM clones in dissected 3^rd instar larval fillets. (I) Quantification of coverage index for control or ban^△1 MARCM clones. For all categories n > 5.

Figure 6

ban expression in epithelial cells is sufficient for scaling growth of dendrites. (A) The ability of transgenic expression of ban directed by a variety of Gal4 drivers to rescue dendrite growth and larval growth defects of ban mutants was assayed at 96 hr AEL. (B) Quantification of rescue activity of Gal4 drivers at 96 hr AEL. n > 10 neurons for each genotype. Black asterisks denote p values relative to wt and red asterisks denote p values relative to ban mutants. (C) Quantification of activity of heat-shock-induced (HS) ban expression to rescue dendrite coverage defects of ban mutants at 96 hr AEL. n > 6 neurons for each category. (D-F) ban levels in epithelial cells affect dendrite growth. Dendrites of class IV ddaC neurons were visualized by a 3x ppk-eGFP reporter for control (3x ppk/+; arm-gal4/+) larvae or larvae in which epithelial cells overexpressed ban (“arm>ban” = 3x ppk/+; arm-gal4/UAS-ban). Due to the temperature-sensitive nature of Gal4, raising larvae at higher temperatures (29°C vs. 22°C) results in higher levels of Gal4 activity, and therefore higher levels of ban expression in epithelial cells. Scale bar is 50 μm.

Figure 7

ban influences dendrite distribution over the body wall epithelium. (A) Positively labeled homozygous mutant ban^△1 epithelial clones (green) were generated in a heterozygous background. PNS dendrites are labeled by HRP immunoreactivity (magenta). (B) Distribution of dendrites over the region directly below epithelial nuclei (denoted by asterisks) was monitored for wt or ban mutant clones. Instances in which PNS dendrites cross this region are color-coded magenta and instances in which this region is devoid of dendrites are color-coded green. Note that dendrites cross this domain in a higher proportion of ban mutant epithelial cells than in neighboring heterozygous cells. (C) Quantification of dendrite growth into the domain shadowed by the epithelial nucleus of control (Frt 2A) or ban mutant epithelial clones. Average number of events per 100 clones of each genotype is shown. n > 500 cells for each genotype.

Figure 8

ban regulates neuronal Akt to ensure dendrite scaling. (A) Akt levels and activity monitored via western blotting of wt or ban larval fillet lysates. (B and C) Expression and activity of Akt in the larval PNS. wt or ban mutant 3^rd instar larvae were immunostained with antibodies to Akt / phospho-Akt (magenta) and HRP (green) to label sensory neurons and their processes. The bracket marks the position of PNS neurons and the arrowhead marks of the class IV neuron ddaC. Genotypes are indicated above the images; twi>ban denotes twist-Gal4, UAS-ban. (D-J) Akt activity influences dendrite scaling in class IV neurons. Dendrite morphology for representative neurons of the following categories: control neuron, wt (D), class IV neuron overexpressing Akt, ppk>Akt, (E), Akt(RNAi) in a class IV neuron, ppk>Akt(RNAi) (F), Akt(RNAi) in a class IV neuron of a ban mutant larva, ppk>Akt(RNAi); ban^△1 (G), overexpression of Pten in a class IV neuron, ppk>Pten (H), or overexpression of Pten in a class IV neuron of a ban mutant larva, ppk>Pten; ban (I). (J) Coverage index for class IV neurons of the indicated genotypes. Asterisks denote p values relative to color-matched controls. n > 10 for each genotype. (K-L) Akt is required for the exuberant dendrite invasion activity in ban mutants. Class IV neurons in “ban^△1 + RNAi control” (genotype: elav-Gal4/+; UAS-Dcr-2/+; ban^△1, ppk-EGFP/ban^△1, ppk-EGFP) and “ban^△1 + neuronal Akt(RNAi)” (genotype: elav-Gal4/+; UAS-Dcr-2/+; ban^△1, ppk-EGFP, UAS-Akt-RNAi/ban^△1, ppk-EGFP) larvae were ablated at 48 hr AEL and imaged 48 hr post-ablation. Representative images are shown (K). Hatched red lines demarcate the dorsal midline and segment boundaries. (L) Quantification of the invasion index for the “ban^△1 + RNAi control” larvae (Control) and “ban^△1 + neuronal Akt(RNAi) larvae (Akt(RNAi)) from (K). n = 4 ablated/imaged neurons for each genotype.