Research Article

Dendritic growth gated by a steroid hormone receptor underlies increases in activity in the developing Drosophila locomotor system

Maarten F. Zwart, Owen Randlett, Jan Felix Evers, and Matthias Landgraf
PNAS October 1, 2013 110 (40) E3878-E3887; https://doi.org/10.1073/pnas.1311711110

  1. Edited by Barry Ganetzky, University of Wisconsin–Madison, Madison, WI, and approved August 26, 2013 (received for review June 20, 2013)

Significance

Why do bigger animals have bigger brains, and how do they get them? We find in Drosophila larvae that motoneuron dendrites, the branched structures receiving information from other neurons, grow as the animal gets bigger, and that this is regulated on a cell-by-cell basis by a specific isoform of a steroid hormone receptor, whose functions were unknown. As these dendrites enlarge, they form more connections with presynaptic partners, leading to greater levels of neuronal activity. We propose that these nerve cells increase their activity to compensate for the demands of a bigger body.

Abstract

As animals grow, their nervous systems also increase in size. How growth in the central nervous system is regulated and its functional consequences are incompletely understood. We explored these questions, using the larval Drosophila locomotor system as a model. In the periphery, at neuromuscular junctions, motoneurons are known to enlarge their presynaptic axon terminals in size and strength, thereby compensating for reductions in muscle excitability that are associated with increases in muscle size. Here, we studied how motoneurons change in the central nervous system during periods of animal growth. We find that within the central nervous system motoneurons also enlarge their postsynaptic dendritic arbors, by the net addition of branches, and that these scale with overall animal size. This dendritic growth is gated on a cell-by-cell basis by a specific isoform of the steroid hormone receptor ecdysone receptor-B2, for which functions have thus far remained elusive. The dendritic growth is accompanied by synaptic strengthening and results in increased neuronal activity. Electrical properties of these neurons, however, are independent of ecdysone receptor-B2 regulation. We propose that these structural dendritic changes in the central nervous system, which regulate neuronal activity, constitute an additional part of the adaptive response of the locomotor system to increases in body and muscle size as the animal grows.

How brain size and neuron size are controlled is not clearly understood. Brain size correlates with body size both between and within different species of animals, and as animals develop and grow their nervous systems also enlarge (1, 2). Differences in brain size are not just due to cell number, but also to the extent of the arborizations that neurons make. For example, both within and between closely related species of mammals, the length and complexity of dendritic arborizations in ganglia of the autonomic nervous system correlate with the size of peripheral target tissues (3, 4). Moreover, the number of primary dendritic branches per neuron correlates with the number of preganglionic neurons providing synaptic input, and this in turn parallels activity levels (3, 5). How the growth of neuronal arbors in the central nervous system is regulated is poorly understood. D’Arcy Thompson, in his seminal work On Growth and Form, suggested that “the ganglion cells […] continue to grow, and their size becomes, therefore, a function of the duration of life” (1). Here, we investigated this notion of cellular growth. Specifically, we asked whether and how the growth of central dendritic processes is regulated as animals increase in size, and we studied the physiological consequences of such structural growth to neuronal function.

As a model system, we work with the Drosophila larva, which grows ∼100-fold in surface area within several days (6⇓–8). Its well-characterized neuromuscular system offers unprecedented access to ask how the nervous system changes during periods of rapid animal growth (9). In Drosophila, as in vertebrates, muscles and neuromuscular junctions enlarge with increases in animal body size (8, 10⇓–12). Enlarging muscle size has important physiological consequences, namely concomitant reductions in input resistance, which means that greater levels of synaptic input are required to evoke effective muscle contractions. At the Drosophila neuromuscular junction, two principal adjustment mechanisms are thought to compensate for size-related changes in muscle physiology. First, the neuromuscular junction grows proportionately with the postsynaptic muscle (8), regulated by the balance of factors that promote and constrain presynaptic terminal growth (13⇓⇓⇓⇓⇓⇓–20). Second, synaptic strength is enhanced by increasing the number of release sites per presynaptic bouton (21) and glutamate receptor-containing densities at postsynaptic specializations (22). Together, these mechanisms maintain an approximately constant amplitude of evoked postsynaptic potentials upon experimental nerve stimulation as the postsynaptic muscle enlarges (23).

In contrast to this wealth of evidence on adjustment processes in the periphery, we still know very little about possible contributions via structural and functional plasticity in the central nervous system. To this end, we examined the development of identified motoneurons in the central nervous system. We find that as animals and their body wall muscles increase in size the central dendritic arbors of motoneurons follow suit, scaling with body size. Those increases in dendritic extent and complexity are associated with increases in the number of synaptic inputs onto motoneurons and translate into higher activity levels, extending the duration of action potential bursts. Interestingly, we find that this motoneuronal growth is regulated on a cell-by-cell basis. It depends on a particular isoform of the steroid hormone receptor ecdysone receptor-B2 (EcR-B2), for which functional roles had not been discovered until now (24⇓–26). The steroid hormone ecdysone regulates transitions between developmental stages and, via other ecdysone receptor isoforms, also triggers remodelling of neurons and nonneuronal tissues during metamorphosis (27). Analogously, in vertebrates, steroid hormones also act as important regulators of neuronal remodelling, for example during puberty in mammals (28, 29), effecting seasonal changes in songbirds (30), or, at the cellular level, regulating the growth of sexually dimorphic muscles and their innervating motoneurons (11, 31, 32). In Drosophila, we find that when the ecdysone steroid signaling pathway is selectively inhibited in single motoneurons these cells fail to extend their dendritic arbors and they do not increase the number of synaptic connections as would normally occur during larval stages, despite being embedded in an otherwise normally developing and growing network. Such neurons with experimentally reduced dendritic arbors retain comparatively low levels of neuronal activity and thus reveal the importance of structural dendritic growth for facilitating increases in synaptic connectivity and drive. The spiking response to membrane depolarization, in contrast, is not regulated by EcR-B2 signaling. We therefore propose that EcR-B2–regulated elaboration of central dendritic arbors, as we have characterized here, represents a central mechanism that mirrors and functionally complements well known adjustments of presynaptic terminals in the periphery that occur in response to changes in body size.

Results

Dendritic Arbors of Neurons Grow as the Animal Grows.

To investigate how nervous systems adapt to dramatic changes in body size, we analyzed how motoneurons change structurally within the central nervous system during a period of rapid growth, namely within the first 48 h after larval hatching (ALH). We visualized single RP2 and aCC motoneurons using a genetic labeling strategy that we have developed (33), which targets gene expression to only a few of the 36 RP2 and aCC motoneurons per nerve cord (one RP2 and one aCC neuron are present in most half-segments). Focusing on RP2 neurons, we imaged these fluorescently labeled cells in acutely dissected nerve cords from precisely staged specimens, from the point of larval hatching (first instar stage) to 48 h ALH (early third instar larval stage) and digitally reconstructed their dendritic arbors using custom software (34, 35) so as to extract accurate quantitative morphometry data about these complex branched cells (Fig. 1 A–D).

Fig. 1.
Fig. 1.

Motoneuron dendrites grow in proportion to larval surface area. (A) Drosophila larvae at 0, 24, and 48 h ALH. (B–D) Projections of confocal Z-stacks of representative RP2 motoneurons at 0, 24, and 48 h ALH, with 3D reconstructions superimposed in B′–D′. (E) Graph detailing mean total dendritic length (in micrometers) of RP2 motoneurons ± SEM at different time points ALH. (F) Correlation between larval surface area and total dendritic length. Each point represents the mean ± SEM. Pearson r shows Pearson product-moment correlation coefficient, with P value indicating the statistical significance of correlation. (G) Sholl analysis, depicting sum of dendritic lengths (mean ± SEM) as a function of the distance from the primary neurite (axon) at 0 h (black), 7 h (blue), 24 h (cyan), 31 h (green), and 48 h ALH (magenta). For all data points n ≥ 5. (Scale bar in A, 0.5 mm and in D’, 5 µm.)

We find that the total dendritic length of these cells increases 3.7-fold between 0 h and 48 h ALH, from 268.9 ± 20.9 µm (± SEM, n = 5) at hatching to 993.4 ± 69.4 µm (n = 5) at 48 h ALH (Fig. 1E). The growth of dendrites increases during the second larval instar (Fig. 1E) and it mirrors the growth of the larval cuticle surface area, a proxy measure for body and body wall muscle size (6). Indeed, these two measures are highly correlated (Pearson r = 0.9944, P < 0.0001) and dendritic length increases proportionally with increases in larval surface area (Fig. 1F).

We wondered whether there were patterns to the addition of new branches during normal dendritic growth. We find that the average length of dendritic segments does not change over developmental time and that dendritic arbors increase by the addition of new segments. To determine the distribution of segments within trees, we applied a Sholl sphere statistical method (36). This shows that most segment additions occur at progressively distal parts of arbors, yet at the same time the relative distribution of segments within arbors is maintained, so that at all observed stages dendritic trees have most of their length concentrated roughly halfway along their expanse with respect to the primary neurite (Fig. 1G).

A Specific Isoform of the Ecdysone Receptor Regulates Dendritic Growth.

Next, we asked how this dendritic growth might be regulated. The characteristic change in the rate at which dendritic arbor size increases during the second larval instar stage led us to investigate the ecdysone signaling pathway, which is important for animal growth (37) and critical for the progression through these developmental transitions (24, 38). The Drosophila ecdysone receptor (EcR) has three isoforms, A, B1, and B2. During larval life, isoforms A and B1 are known to be expressed only immediately before pupariation (39). For the EcR-B2 isoform, in contrast, its expression profile and role have remained uncharacterized (24, 25, 39, 40). To determine which EcR isoforms are expressed in the ventral nerve cord during stages of rapid dendritic growth, namely during the second larval instar, from 24 to 48 h ALH, we took advantage of well-characterized antibodies that either recognize all three receptor isoforms (α-EcR-common) or that are specific to the A or B1 isoform (40). No B2 isoform-specific antibody is available, because the EcR-B2 differs from other isoforms by only its N-terminal 17 amino acids (25, 40). Confirming previous reports (39), we do not detect expression of isoforms A and B1 in the first or second instar larva (Fig. S1A, B, D, and E). In contrast, the antibody recognizing all three isoforms shows clear nuclear staining in many cells in the nervous system at these stages, including the aCC and RP2 motoneurons (Fig. 2A and Fig. S1 C and F). This suggests that the B2 isoform is present during early larval stages. To confirm that this nuclear staining represents the EcR-B2 isoform, we analyzed nerve cords from EcR-B−/− mutant larvae that lacked the B2 epitope and nerve cords containing individual cells that overexpress EcR-B2. In the EcR-B−/− nerve cords, the α-EcR-common signal is severely reduced, if not entirely absent, whereas in cells that overexpress EcR-B2 the staining intensity is increased (Fig. 2 B and C). These findings strongly suggest that the EcR-B2 isoform is the only EcR isoform present during early larval stages in many neurons, including the aCC and RP2 motoneurons.

Fig. 2.
Fig. 2.

EcR-B2 is required for RP2 dendritic growth during larval development. (A–C′) Z-projections of confocal image stacks detailing immunofluorescence detection of EcR expression (green) in second instar larval nerve cords in which also single aCC and RP2 motoneurons have been labeled by RFP expression (magenta). (A, A′) Control nerve cord, (B, B′) EcR-B−/− mutant central nervous system, (C, C′) control nerve cord in which single aCC and RP2 motoneurons overexpress EcR-B2 (magenta). (D–F) Three-dimensional reconstructions of dendritic arbors, color-coded according to the distance from the origin of the arbor, showing a control RP2 neuron (D), an RP2 neuron expressing the dominant-negative EcR-B2W650A (DN-EcR-B2), and an RP2 neuron coexpressing DN-EcR-B2 and full-length wild-type EcR-B2. Cell bodies have been omitted for the sake of clarity, but primary neurites/proximal axons are shown, color-coded dark purple. (G) Total dendritic length of control RP2 motoneurons at different developmental time points ALH, mean ± SEM; control (blue), cells expressing DN-EcR-B2 (red), overexpressing wild-type EcR-B2 (green), and coexpressing DN-EcR-B2 and wild-type EcR-B2 (yellow). Mean ± SEM, n ≥ 5, asterisk indicates P < 0.05. (H) Sholl analysis, depicting sum of dendritic lengths as a function of the distance from the primary neurite (axon), of control cells at 24 h (dashed blue line) and 48 h ALH (solid blue line), cells expressing DN-EcR-B2 at 48 h ALH (red), and cells coexpressing DN-EcR-B2 and EcR-B2 at 48 h ALH (yellow). n ≥ 5, asterisk indicates P < 0.05.

Next, we investigated whether EcR-B2 is required for dendritic growth. Because there are no mutant alleles that exclusively compromise EcR-B2 function we studied EcR-B−/− larvae homozygous mutant for both the EcR-B1 and EcR-B2 isoforms. At 48 h ALH EcR-B−/− mutant animals are overall reduced in size and RP2 neurons have smaller dendritic trees compared with heterozygote controls (Fig. S2A, B, and D). Selective reinstatement of EcR-B2 in RP2 motoneurons within such mutant larvae, through targeted expression EcR-B2, partially rescues the dendritic phenotype (Fig. S2 C–E). However, EcR-B−/− mutant animals have several developmental defects in addition to reduced larval growth, such as an inability to molt (24). To selectively compromise EcR-B2 function in specific neurons without otherwise affecting nervous system or animal development, we took advantage of a dominant-negative form of EcR-B2 (UAS-EcR-B2W650A), which is unable to bind ligand (25, 26). We find that targeting expression of the dominant-negative EcR-B2W650A to individual RP2 motoneurons from late embryonic stages onward severely inhibits dendritic growth during the second instar stage compared with controls (Fig. 2 D–G). This effect on dendritic growth is also manifest when UAS-EcR-B2W650A is targeted to another motoneuron, aCC (Fig. S3). Detailed analysis of RP2 dendritic arbors at 48 h ALH shows that expression of the dominant-negative EcR-B2W650A significantly decreases the number of branches, overall tree volume, and tree density compared with controls. Average branch length, however, is not affected, indicating that dendritic growth deficits arise through reduced net addition of branches (Fig. S4 A–D). To determine the distribution of dendritic branches we carried out a Sholl sphere statistical analysis (36). This shows that cells expressing the dominant-negative EcR-B2W650A have most of their dendritic length located approximately halfway along the dendritic tree, as is also characteristic for control neurons (Fig. 2H). Specifically, at 48 h ALH RP2 neurons expressing EcR-B2W650A have dendritic arbors that are comparable to control RP2 neurons at 24 h ALH, both with respect to total dendritic length and distribution of dendritic branches within the arbor (Fig. 2 G and H). Furthermore, the soma, which normally increases in size, does not undergo statistically significant growth between 24 and 48 h ALH (Fig. S5), suggesting that, at least morphologically, RP2 neurons expressing EcR-B2W650A arrest in their second instar state.

We next tested the specificity of action of UAS-EcR-B2W650A. EcR-B2W650A carries a mutation in the ligand-binding domain, and as a result this mutated receptor fails to bind 20E and does not activate transcription (25, 26). However, it still binds its coreceptor, ultraspiracle, and therefore EcR-B2W650A acts as a competitive inhibitor of wild-type EcR (25). Consistent with this prediction, expression of wild-type forms of EcR subunits has no discernible effect on dendritic growth, and coexpression of the wild-type EcR-B2 suppresses UAS-EcR-B2W650A–induced dendritic growth defects, as would be expected from competitive interactions (Fig. 2G).

Taken together, these results show that dendritic growth is regulated cell-by-cell, rather than at the level of the whole tissue. Specifically, in this system EcR-B2 signaling acts cell-autonomously by regulating whether or not a neuron proceeds to enlarge its dendritic arbor during development, irrespective of its surroundings and its partner neurons developing normally.

Neuronal Activity Levels Increase with Neuronal Size and Are Regulated by EcR-B2.

Next, we asked what the functional significance might be for the developmentally regulated increases in dendritic size. To this end, we made electrophysiological recordings from the RP2 and aCC motoneurons. Both neurons innervate dorsal muscles and receive identical synaptic input in our electrophysiological assays, as also reported by Baines et al. (41). We focused on the two developmental time points that are characterized by the greatest difference in dendritic growth during the period of analysis: 24 and 48 h ALH, the early second and third instar larval stages, respectively.

To record endogenous activity patterns, we first used a cell-attached configuration. These neurons display rhythmic bursts of activity, and the number of action potentials per burst fired by aCC increases from the second to the third instar stage, from 18.8 (SEM = 1.8, n = 10) at 24 h ALH to 44.6 spikes per burst (SEM = 6.6, n = 11) at 48 h ALH (Fig. 3 A and C). In line with this, burst duration increases from 0.51 s (SEM = 0.08) at 24 h ALH to 1.08 s (SEM = 0.23, P = 0.03) at 48 h ALH (Fig. 3E), whereas the mean spike frequency within each burst (intraburst spike frequency) does not change (Fig. 3D). Thus, as motoneuron dendritic arbors enlarge, scaling with increases in larval size, the number of action potentials fired per burst of activity also rises. Curiously, the frequency of bursts decreases from 0.45 Hz (SEM = 0.04) at 24 h ALH to 0.26 Hz (SEM = 0.03, P = 0.002) at 48 h ALH (Fig. 3F). We find no statistically significant correlation between the spike number per burst and the frequency of bursts at 48 h ALH (Spearman's ρ = 0.01, P = 0.84) (Fig. 3G). This suggests that the number of action potentials per burst could be set by factors affecting either cell-intrinsic and/or synaptic properties, rather than being a cycle period-dependent property of the network.

Fig. 3.
Fig. 3.

EcR-B2 is required for increases in neuronal activity during the second larval instar. (A and B) Cell-attached recordings of endogenous bursting behavior of control aCC motoneurons (A) and cells expressing the dominant-negative EcR-B2W650A (DN-EcR-B2) (B) at 24 and 48 h ALH. Number of action potentials per burst (C), action potential frequency within each burst (D), burst duration (E), and burst frequency (F) at 24 and 48 h ALH for control cells (blue) and cells expressing DN-EcR-B2 (red). (G) Correlation between the number of action potentials per burst and the burst frequency, with Spearman's rank correlation coefficient and corresponding P value. Mean ± SEM for C–F. Asterisks indicate P < 0.05.

Given the requirement for EcR-B2 signaling for dendritic growth as larvae progress from the second (24 h ALH) to the third instar stage (48 h ALH), we wondered whether EcR-B2 signaling was also necessary for associated increases in neuronal activity. To this end, we targeted EcR-B2W650A expression to aCC and RP2 motoneurons and made cell-attached recordings of endogenous activity patterns at 24 and 48 h ALH. We find that unlike in control neurons the number of action potentials per burst fails to increase during this period when EcR-B2W650A is expressed (control at 24 h ALH: 18.8 spikes per burst, SEM = 1.8, n = 10; at 48 h ALH: 44.6 spikes per burst, SEM = 6.6, n = 11; RN2 > EcR-B2W650A at 24 h ALH: 17.8 spikes per burst, SEM = 1.2, n = 7; at 48 h ALH: 19.6 spikes per burst, SEM = 3.3, n = 8; P < 0.01 for control-experimental comparison at 48 h ALH) (Fig. 3 A–C). The mean spike frequency within bursts remains similar from 24 to 48 h ALH in both control and EcR-B2W650A–expressing motoneurons (Fig. 3D). Consequently, neurons in which EcR-B2 is inhibited retain comparatively short action potential burst durations, unlike control cells, which increase the burst duration from 24 to 48 h ALH (control at 24 h ALH: 0.51 s, SEM = 0.08; at 48 h ALH: 1.1 s, SEM = 0.23; RN2 > EcR-B2W650A at 24 h ALH: 0.32 s, SEM = 0.06; at 48 h ALH: 0.35 s, SEM = 0.05; P < 0.05 for control-experimental comparison at 48 h ALH) (Fig. 3E). The frequency at which these bursts of activity occur decreases similarly in both control and EcR-B2W650A–manipulated neurons from 0.43 Hz (SEM = 0.03) to 0.32 Hz (SEM = 0.05) (Fig. 3F).

These data suggest that EcR-B2 signaling is required in neurons during larval development both for dendritic growth and for associated increases in neuronal activity. Remarkably, targeted interference with EcR-B2 signaling in single motoneurons blocks their normal developmental increase in action potential burst duration, even though the remainder of the network develops normally (i.e., there is no experimentally induced change in burst frequency). This suggests that burst duration is regulated cell-autonomously downstream of EcR-B2.

Spike Response to Membrane Depolarization Is Not Regulated by EcR-B2.

We wondered whether the observed developmental change in neural activity might be caused by changes in the intrinsic excitable properties of these neurons. First, we looked at the spike threshold: the membrane voltage at which the first action potential is fired. To determine this threshold, we depolarized the neurons from −120 to +60 mV over 500 ms using voltage clamp. Space clamp errors cause more distal regions of the cell to be poorly clamped (42) and these neurons therefore fire action potentials in response to the depolarizing ramp. Usually this experiment is done in current clamp, but embryonic and early larval neurons are less well suited for this technique owing to membrane voltage fluctuations that are difficult to control (41, 43). We found that the voltage threshold at which the first action potential is fired does not change significantly between 24 and 48 h ALH (−22.3 mV, SEM = 1.0, n = 9 at 24 h; −23.6 mV, SEM = 0.73, n = 9 at 48 h, P > 0.05), and it is not significantly affected by expression of EcR-B2W650A (Fig. 4A). This suggests that the observed developmental increase in activity is not caused by neurons firing earlier in response to synaptic input.

Fig. 4.
Fig. 4.

EcR-B2 is not required for changes in the electrical excitability of aCC during development. (A) Whole-cell voltage-clamp recording of an aCC motoneuron measuring whole-cell currents in response to a depolarizing ramp from −120 to +60 mV. Graph depicts voltage threshold of control cells (blue) and cells expressing the dominant-negative EcR-B2W650A (DN-EcR-B2) (red) at 24 and 48 h ALH. (B and F) Whole-cell current-clamp recordings of aCC in response to 1-s injections of depolarizing currents of 10, 40, and 80 pA amplitude at 24 and 48 h ALH of control cells (B) and cells expressing DN-EcR-B2. (C) Electrical excitability measured as spikes/action potentials fired per second in response to current injections at 24 h (dashed line) and 48 h ALH (solid line) of control cells (blue) and cells expressing DN-EcR-B2 (red). (D) Action potentials fired per second as a function of membrane potential (Vm) for control cells at 24 h (open blue squares) and 48 h (solid squares). Lines indicate linear regression plots for control cells at 24 h (dashed line) and 48 h (solid line). There is no statistically significant difference between the two regression plots in either slope or intercept (P > 0.05). (E) Action potentials fired per second as a function of Vm for control cells (solid blue squares) and cells expressing DN-EcR-B2 at 48 h ALH (solid red squares). Lines indicate linear regression plots for control cells (blue line) and experimental cells (red line). There is no statistically significant difference between the two regression plots in either slope or intercept (P > 0.05). (F) Input resistance of control aCC motoneurons (blue) and cells expressing DN-EcR-B2 (red) at 24 and 48 h ALH. (G) Cell capacitance of control aCC motoneurons (blue) and cells expressing DN-EcR-B2 (red) at 24 and 48 h ALH. Mean ± SEM, n ≥ 8. Asterisk indicates P < 0.05.

Second, the level of activity could be determined by the neurons’ excitability, separately from the voltage threshold. We therefore measured electrical excitability at 24 and 48 h ALH by injecting 1-s steps of current of increasing magnitude in the presence of an α7-nicotinic acetylcholine receptor antagonist, α-bungarotoxin, which blocks all excitatory input (43, 44). We find that the number of action potentials fired for a given current injection decreases in older animals but that it is not affected by expression of EcR-B2W650A (for 100-pA injection at 24 h ALH: control = 48.0 spikes, SEM = 2.1, n = 10 vs. RN2 > EcR-B2W650A = 48.3 spikes, SEM = 3.2, n = 8; at 48 h ALH: control = 19.2 spikes, SEM = 2.4, n = 11 vs. RN2 > EcR-B2W650A 17.8 spikes, SEM = 4.4, n = 8) (Fig. 4 B and C). To assess membrane excitable properties, we measured the number of action potentials fired at a given membrane potential. We find that this remains constant during these stages and that it is unaffected by expression of the dominant-negative EcR-B2W650A (Fig. 4 D and E). Therefore, changes in voltage-gated ion channels that generate the action potential are not the cause for the decrease in electrical excitability that normally occurs over developmental time, but other factors are responsible. In line with this, we find that input resistance decreases, as would be expected from increases in cell size and therefore leak currents (Fig. 4F). In addition, capacitance increases over developmental time, although less for EcR-B2W650A–expressing neurons than for controls (Fig. 4G and Fig. S5). Despite these differences in capacitance, controls and EcR-B2W650A–expressing neurons have the same efficacy of depolarizing their action potential initiation zone at both 24 and 48 h ALH.

Taken together, these data suggest that, although during normal development we observe a rise in neuronal activity from 24 to 48 h ALH, the neurons actually become intrinsically less excitable. This decrease in intrinsic excitability is not regulated by EcR-B2. Although it is at least partly a consequence of increased cell size, other factors possibly contribute. For example, relative location of the action potential initiation zone could be an important determinant of neuronal excitability. This specialized axonal region has not been located in these cells but in adult Drosophila motoneurons is thought to be positioned more than 100 µm from the soma (45).

EcR-B2–Mediated Increases in Synaptic Connections Cause Developmentally Regulated Elevation in Neuronal Activity.

Having ruled out changes in the intrinsic excitability, we investigated extrinsic causes for the increase in neural activity that occurs between 24 and 48 h ALH, such as elevated levels of synaptic input. To test this hypothesis, we made whole-cell recordings from these motoneurons (voltage-clamped at −60mV) and measured evoked postsynaptic currents (EPSCs). We focused on sustained currents previously shown to initiate firing of action potentials (41). The mean amplitude of synaptic currents increases from 24 to 48 h ALH and is unaffected by expression of EcR-B2W650A (at 24 h: control = 371.3 pA, SEM = 44.9, n = 13 vs. RN2 > EcR-B2W650A = 456.7 pA, SEM = 44.7, n = 10; at 48 h: control = 668.7 pA, SEM = 61.7, n = 14 vs. RN2 > EcR-B2W650A = 626.3 pA, SEM = 60.9, n = 8, P < 0.05 between time points; P > 0.05 between genotypes) (Fig. 5 A and B). However, in 48-h–old animals expression of the dominant-negative EcR-B2W650A leads to a reduction of the total synaptic drive, as measured by the integral of synaptic current (RN2 > EcR-B2W650A: 147 nA·ms, SEM = 17; control: 240 nA·ms, SEM = 34, P = 0.02) (Fig. 5 C and D). This reduction of the total synaptic drive is most likely the reason, or at least part thereof, as to why the number of action potentials per burst is reduced when EcR-B2 is inhibited in these motoneurons. These results suggest that the changes in activity that we observe during normal development, and that are regulated by EcR-B2 signaling, must be caused by enhanced levels of synaptic input.

Fig. 5.
Fig. 5.

Synaptic drive is enhanced through EcR-B2–dependent developmental increase in innervation. (A) Whole-cell voltage-clamp recordings of endogenous evoked excitatory postsynaptic currents (EPSCs) of control aCC and RP2 motoneurons and cells expressing the dominant-negative EcR-B2W650A (DN-EcR-B2) at 24 and 48 h ALH (holding potential −60 mV). (B) Absolute amplitude of EPSCs of control cells (blue) and cells expressing DN-EcR-B2 (red) at 24 and 48 h ALH. (C) Overlay of 20 traces of whole-cell voltage-clamp recordings of endogenous evoked excitatory synaptic currents, of a control cell (blue) and a cell expressing DN-EcR-B2 (red) at 48 h ALH. (D) Integral of EPSCs for control cells and cells expressing DN-EcR-B2 at 48 h ALH. (E) Whole-cell voltage-clamp recordings of endogenous spontaneous synaptic currents, miniature synaptic postsynaptic currents (mEPSCs), of control aCC and RP2 motoneurons and cells expressing DN-EcR-B2 at 24 and 48 h ALH (holding potential −65 mV). Frequency (F) and absolute amplitude (G) of mEPSCs of control cells (blue) and cells expressing DN-EcR-B2 (red) at 24 and 48 h ALH. (H) Frequency histogram of absolute amplitude of mEPSCs at 48 h ALH. Mean ± SEM, n ≥ 11 for A–D, n ≥ 6 for E–H. Asterisk indicates P < 0.05.

Two principal mechanisms could underlie this synaptic strengthening: increased presynaptic neurotransmitter release or elevated postsynaptic sensitivity to neurotransmitter. To distinguish between these, we generated whole-cell patch-clamp recordings from the motoneurons, voltage-clamped at −65mV, in the presence of TTX. TTX suppresses evoked release of neurotransmitter, leaving only that which is released by the spontaneous fusion of single or a few vesicles. In controls, we observed a rise in the mean frequency of miniature excitatory postsynaptic currents (mEPSCs) from 24 to 48 h ALH: 1.6 Hz at 24 h (SEM = 0.22) vs. 4.3 Hz at 48 h ALH (SEM = 0.24, P = 0.008) (Fig. 5 E and F). This normally occurring rise in mEPSC frequency is blocked by expression of the dominant-negative EcR-B2W650A (RN2 > EcR-B2W650A at 24 h ALH: 1.6 Hz, SEM = 0.22, n = 6; at 48 h ALH: 2.2 Hz, SEM = 0.45, n = 7; P > 0.05) (Fig. 5 E and F). The likely cause is a stagnation of the number of release sites that form on these dendritic arbors between 24 and 48 h ALH, mirroring the lack of dendritic growth during the same period when EcR-B2 signaling is compromised.

The amplitude of mEPSCs does not change from 24 to 48 h ALH during normal development, although it increases in cells that express EcR-B2W650A (control at 24 h ALH: 12.9 pA, SEM = 0.46; at 48 h ALH: 10.9 pA, SEM = 1.1, P > 0.05; RN2 > EcR-B2W650A at 24 h ALH: 15 pA, SEM = 2.0; at 48 h ALH: 24.2 pA, SEM = 2.7, P < 0.05) (Fig. 5 G and H). This EcR-B2W650A–dependent increase in the amplitude of mEPSCs could be a biophysical consequence of reduced membrane area in neurons that express EcR-B2W650A relative to the comparatively larger control cells. Alternatively, it might reflect an adjustment of the manipulated cells to compensate for reductions in (anticipated) synaptic drive, a process that resembles synaptic scaling (46⇓–48).

In summary, we find that the EcR-B2 isoform is required in central neurons during postembryonic larval stages to increase synaptic drive, likely through the addition of presynaptic sites and mediated by EcR-B2–regulated dendritic growth, which results in higher levels of neuronal activity.

Discussion

As animals develop and increase in size, many of their neurons also enlarge. We investigated this phenomenon and found that, at least in this system, neuronal growth is regulated so that neurons individually take the decision of whether or not to enlarge their central dendritic arbors. This gating of dendritic growth is regulated by a specific isoform of the steroid hormone receptor EcR-B2. Using electrophysiological recordings we determined the functional consequences of this neuronal growth that normally occurs during development. Motoneurons in which EcR-B2 signaling has been compromised retain smaller dendritic arbors with fewer synaptic connections than their control counterparts, mimicking younger and smaller neurons, despite being situated in an otherwise unmanipulated ganglion. The consequences are reduced synaptic drive and shorter activity periods. The implications of these observations are twofold: First, neurons can deploy structural changes in their dendritic trees as a central mechanism with which to regulate and adjust levels of neuronal activity; second, in terms of connectivity, the size of the postsynaptic dendritic arbor seems to be decisive in determining the number of connections that neurons form among available presynaptic terminals.

Motoneuron Dendritic Growth in the Central Nervous System Scales with Increases in Animal Body Size.

Increases in body or organ size are normally accompanied by matching changes in innervation, required to maintain appropriate neuronal control. One of the best-studied examples is the neuromuscular junction where increases in muscle size lead to biophysical changes in muscle physiology, which are compensated for by a matching enlargement of the neuromuscular junction (8, 10, 11, 49). The neural and cellular mechanisms that regulate these homeostatic adjustments have been studied in detail (for reviews see refs. 50 and 51). Here, we have identified a potential additional, central mechanism associated with adjustment to growth in the neuromuscular system: In the central nervous system, motoneurons also enlarge their postsynaptic dendritic arbors as animals increase in body size; this leads to increased synaptic drive and thus prolonged periods of bursting activity.

Our morphometric quantifications of dendritic arbors during larval stages, show that motoneuron dendritic trees increase their overall dendritic length proportionately to body size. The growth of these arbors occurs by addition of new dendritic segments that “fill in” existing territory, as well as at the perimeter of the tree, thus widening its reach. Such scaling growth of dendritic arbors in relation to body size is a widespread phenomenon and has previously been observed in different types of nerve cells, including Purkinje, pyramidal, olfactory mitral cells and sympathetic ganglionic neurons (3, 4, 52, 53).

Ecdysone Receptor Isoform B2 Is Required Cell-Autonomously for Normal Dendritic Growth.

An interesting discovery of this study is that the growth that normally occurs during larval stages is regulated cell-autonomously. Specifically, we established that the B2 isoform of the ecdysone steroid hormone receptor is required cell-autonomously in motoneurons for normal dendritic growth. Expression of a dominant-negative form of EcR-B2, UAS-EcR-B2W650A, in single cells prevents the characteristic increase in motoneuron dendritic arbor size during the second larval instar stage and seems to arrest neural arbors structurally at a young larval stage, despite being embedded in an otherwise normally developing ganglion. Electrical excitability of aCC and RP2 motoneurons, however, are not affected by expression of EcR-B2W650A. We found EcR-B2 to be the only isoform expressed in the larval nerve cord during early larval stages, in agreement with and complementary to previously published data (39). Although different functions have been ascribed to the other two EcR isoforms, A and B1 (e.g., refs. 24 and 54⇓⇓–57), the role of the B2 isoform had until now remained unknown (26). Here, we uncovered an important role for the EcR-B2 isoform in nervous system development, namely to permit growth in larval stages. We interpret the role of EcR-B2 as being a permissive factor for dendritic growth for two reasons. First, the pattern of dendritic growth during larval stages does not follow ecdysteroid titers (38) but may be exponential. Second, precocious or overexpression of the wild-type form of EcR-B2 does not cause abnormal dendritic growth. Because the nervous system is one of the most metabolically expensive tissues (58), it is conceivable that gating the decision through EcR-B2 on whether or not neurons grow may provide a strategy to synchronize neural growth with the growth of the animal as a whole, as it progresses from one developmental stage to the next. Indeed, production of ecdysteroids in the Drosophila larva is under nutritional control (59).

It is likely that other signaling pathways determine the extent to which neurons grow. For example, in rat superior cervical ganglion cells dendritic growth correlates with peripheral target size (60) and NGF has been implicated (61, 62). In Drosophila, three neurotrophic factors have been identified, expressed in subsets of body-wall muscles at embryonic stages, although none has thus far been reported to be expressed in the dorsal musculature, whose innervating aCC and RP2 motoneurons we analyzed here (63). Other muscle and associated glia-derived retrograde regulators of neuromuscular junctions include the TGF-β homologs Glass bottom boat and Maverick and the Wnt family member Wingless (15, 18, 19, 64⇓–66). It is conceivable that these could regulate the growth of motoneuron dendritic arbors in synchrony with that of presynaptic axon terminals. Indeed, we find that, in addition to its role in regulating dendritic growth, EcR-B2 is also required for normal neuromuscular junction growth, suggesting that EcR-B2 itself regulates the development of both pre- and postsynaptic compartments (Fig. S6).

Dendritic Growth as a Central Mechanism to Adjust Neuronal Activity Levels.

As the Drosophila larva grows around 100-fold in surface area with matching increases in body wall muscle size, appropriate levels of muscle depolarization have to be maintained. The larval body-wall muscles are virtually isopotential (67), and as their input resistance goes down with increasing muscle size, presynaptic output at the neuromuscular junction increases in a compensatory fashion: By adjusting both terminal size and synaptic strength the amplitude of postsynaptic responses in the muscle is maintained (23, 68⇓⇓–71). Similarly, at the growing vertebrate neuromuscular junction, motor endplates expand as muscles enlarge (12), enhancing neurotransmitter release (49).

Our findings show that in addition to the well-characterized regulation of neuromuscular junction strength, in the central nervous system, motoneurons also adjust the size of their dendritic terminals. We previously demonstrated that during the initial assembly of the locomotor network in the embryo neurons deploy their dendritic arbors as structural homeostatic devices, adjusting their extent to compensate for naturally occurring variations in the density of synaptic partner terminals (72). In this study, we have shown that in subsequent larval stages, when animals grow rapidly, extension and elaboration of dendritic arbors leads to greater numbers of presynaptic inputs and thus increased synaptic drive. Specifically, with progression from the second to the third larval instar the duration of action potential bursts approximately doubles. Longer action potential burst periods might increase and potentially prolong muscle contractions; they could also enhance facilitation (73, 74). Furthermore, the time course of excitatory junctional potentials (EJPs) at the Drosophila neuromuscular junction changes as muscles enlarge (23): Time constants describing both the rise and fall of EJPs increase, which, taking into account the bursting input the muscle receives, will result in a larger envelope of depolarization as the animals grows. It is therefore conceivable that the increased number of action potentials fired per burst that we have observed affects the strength of the neuromuscular synapse by enhancing both the process of facilitation and the envelope of depolarization.

EcR-B2–Mediated Dendritic Growth Increases Synaptic Drive Through Additional Synaptic Contacts.

Having identified EcR-B2 as a regulator of dendritic growth of motoneurons, we were able to investigate how dendritic growth relates to synaptic drive in this system. During normal development, motoneurons increase their dendritic arbor proportionally to animal body size. A biophysical consequence of increased neuronal size is increased capacitance and decreased input resistance, both of which reduce the cell’s intrinsic excitability (75). Indeed, we find that motoneurons in third instar larvae (48 h ALH) have larger dendritic arbors and are less excitable than smaller cells of younger, second instar (24 h ALH) animals. We also found that increases in dendritic arbor size are accompanied by increases in the frequency of spontaneous mEPSPs (Fig. 5), suggesting that the dendritic growth during larval development facilitates the addition of synapses.

We wondered how these changes in synaptic input, from the second to the third larval instar, might lead to extended bursting periods. Most likely, this requires the addition of synapses from new premotor partner neurons. Interestingly, at 48 h ALH EcR-B2W650A–expressing motoneurons, despite being located within an otherwise unmanipulated nervous system with its third instar complexity and density of presynaptic release sites, are indistinguishable from younger neurons in a younger ganglion, in terms of dendritic arbor size, distribution of dendritic branches in the neuropil, number of synaptic sites on these, and activity patterns (Figs. 2, 3, and 5). This suggests that the size and geography of these dendritic arbors is decisive in determining their connectivity. Because we find increases in the duration of action potential bursts over developmental time, it is likely that as they grow aCC and RP2 motoneuron arbors establish new presynaptic contacts from additional interneurons, some of which may be in adjacent segments. These could include segmental homologs of those with whom they already form connections more proximally at earlier stages. Such a scenario could extend the duration of the synaptic drive, as we observe: As each wave of activity passes through the nerve cord during locomotion cycles, synapses in adjacent segments would have different timings that when combined on one dendritic arbor could lead to prolonged periods of action potential bursts. Indeed, tentative evidence has shown that larval Drosophila motoneurons change their connectivity qualitatively, in that neurons begin to show inhibitory responses during larval development (43, 76).

Comparable strategies have been documented in other systems. For instance, the substantia nigra compacta dopaminergic neurons change their dendritic architecture to alter the number and identity of synapses they receive and thereby also their functional properties within the network (77). In the case of the aCC and RP2 motoneurons in Drosophila that we have studied here, the identity of their excitatory presynaptic partners has not yet been characterized beyond being cholinergic (43) and so at this point cannot be resolved conclusively.

EcR-B2 Regulates the Development of Structural Attributes but Not of Intrinsic Physiological Properties.

We found that the electrical excitability of aCC/RP2 is not affected by expression of EcR-B2W650A (Fig. 4). The excitability of a cell is determined by the input resistance, the sum of all leak currents at rest, and the voltage-sensitive conductances that generate the action potential. The input resistance of a cell is normally inversely proportional to its size (75). Consistent with this notion, we find that during normal development the electrical excitability of these neurons decreases during the second larval instar stage, as cells increase in size. Moreover, we do not record changes in the relationship between membrane potential and action potential firing, suggesting there is no net change in the voltage-sensitive conductances that generate the action potential, and this aspect is not affected by EcR-B2W650A expression (Fig. 4 D and E). However, the inverse relationship between cell size and excitability no longer holds for cells expressing EcR-B2W650A: At 48 h ALH, the intrinsic excitability of these comparatively small cells, which are reduced both in dendritic arbor and soma size (Fig. 2 and Fig. S5), is similar to that of their age-matched, larger, control counterparts (Fig. 4C). These findings are compatible with at least two scenarios. First, it is possible that the location of the action potential initiation zone, a key determinant of excitability of these neurons, changes from the second to the third larval instar stage. For example, the positioning of the action potential initiation zone relative to the proximal primary neurite, which integrates dendritic currents, may change as nerve cords enlarge, and this could be independent of EcR-B2 signaling in individual motoneurons. However, in agreement with an earlier study, we find that the intrinsic excitability of aCC and RP2 motoneurons strongly correlates with the amplitude of synaptic input (78). In this model, EcR-B2W650A–expressing neurons would undergo homeostatic adjustment to remain within the normal range of neuronal activity. Given that we did not detect changes in the voltage-sensitive conductances that generate the action potential, but measured a reduction in membrane resistance over developmental time, leak channels could be involved.

Materials and Methods

Fly Strains.

Wild-type and transgenic strains were maintained on standard yeast–agar–cornmeal medium at 25 °C. The following fly strains were used: Oregon-R (Bloomington Stock Center, Indiana University), w1118, eveRN2-Flippase, UAS-myr::mRFP, UAS-Flp, tubulin-FRT-CD2-FRT-GAL4 (79) and w1118, eveRN2-GAL4 (RN2-GAL4) (80), UAS-myr-mRFP1, UAS-Flp, tubulin84B-FRT-CD2-FRT-GAL4 (79). The RN2-GAL4 line expresses in aCC and RP2 motoneurons in the embryo, but not during larval stages. In combination with the FLPase-gated tubulin84B-FRT-CD2-FRT-GAL4, a subset of these neurons maintains expression at high levels. UAS-EcR-B2W650A, UAS-EcR-B2, UAS-EcR-AW650A, UAS-EcR-A, UAS-EcR-B1W650A, UAS-EcR-B1 (25, 26) (Bloomington Stock Center, Indiana University). Homozygous lethal insertions were kept over CyO, Dfd-GMR-YFP or TM6b, Sb, Dfd-GMR-YFP balancer chromosomes (Bloomington Stock Center, Indiana University).

Staging and Dissection for Structural Analysis.

Flies were allowed to lay eggs on apple juice-based agar medium overnight at 25 °C. Hatched larvae were removed, after which the remaining eggs were incubated at 25 °C for a further 2 h. Larvae hatched during this period (designated 0 h ALH) were then collected and either dissected in Sorensen’s saline (pH 7.2, 0.075 M) or allowed to develop further in a yeast paste until the appropriate stage was reached (designated relative to the first time point; therefore, each sample, although containing larvae from a 2-h developmental time window, was designated as if from a single time point). A fine hypodermic needle (30 1/2 G; Microlance) was used as a scalpel to cut off the anterior end of each larva, allowing gut, fat body, and trachea to be removed. The ventral nerve chord and brain lobes, extruded with viscera upon decapitation, were dissected in Sorensen’s saline and positioned, dorsal side up, on a cover glass coated with poly-l-lysine (Sigma-Aldrich). The cover glass was placed on a microscope slide using glycerol as an adhesive. A clean cover glass was placed on top of the preparation, with strips of double-sided sticky tape as spacers positioned along the edges.

Confocal Imaging, Dendritic Reconstruction, and Image Processing.

RFP-expressing neurons were imaged immediately after dissection without fixation with a Yokagawa CSU-22 spinning disk confocal field scanner mounted on an Olympus BX51WI microscope, using a 60×/1.2 N.A. Olympus water immersion objective. Images were acquired with a voxel size of 0.2 × 0.2 × 0.3 µm. Dendritic trees were digitally reconstructed using Amira Resolve RT 4.1 (Visualization Sciences Group and Zuse Institute), supplemented with statistical algorithms developed by J.F.E. (34, 35), and images were processed using Amira and ImageJ (National Institutes of Health).

Larval Surface Area Calculation.

Larval length (l) and diameter (d) were measured using ImageJ. Larvae were approximated as ellipsoids and surface area (SA) was calculated using the following equation (81):Embedded Image

with P = 1.6075.

Electrophysiology.

Larvae were dissected and central neurons accessed as described previously (43). Briefly, the larval central nervous system was removed and pinned onto a Sylgard-coated dish using fine wire (0.001 tungsten 99.95% wire; California Fine Wire Company). A small section of the glial sheath surrounding the ventral nerve cord between segments A1–A3 was ruptured using protease (0.1–1% Protease XIV; Sigma-Aldrich) dissolved in external saline (discussed below) to expose motoneuron cell bodies underneath. The preparation was viewed with a 60×/0.9 N.A. water-dipping objective using DIC microscopy (BX50WI microscope; Olympus). Whole-cell and cell-attached recordings were performed using standard thick-walled borosilicate electrodes (GC100TF-10; Harvard), fire-polished to resistances of 7–10 MΩ for third instar and 10–15 MΩ for second instar preparations. Recordings were made using an Axopatch-1D amplifier and digitized using a Digidata 1322A. Traces were recorded using pClamp 10 (all from Molecular Devices), digitized at 20 kHz and filtered at 2 kHz. After breakthrough, currents were measured for a period ranging from 2 min to 4 min; cells begin to pass increased leak currents after this time (41). Cell-attached recordings were performed until a deterioration in the signal-to-noise ratio was detected. Data were analyzed using Clampfit 10 (Molecular Devices) and Spike2 (Cambridge Electronic Design). For miniature current recordings, all measurements were verified by eye. External saline consisted of 135 mM NaCl, 5 mM KCl, 4 mM MgCl2⋅6H2O, 2 mM CaCl2⋅2H2O, 5 mM N-Tris[hydroxymethyl]methyl-2-amonoethanesulfonic acid, and 36 mM sucrose (pH 7.15), supplemented with 0.1–0.2 µM TTX (voltage-sensitive sodium channel blocker; Abcam) to record spontaneous (miniature) currents or 1 µM α-bungarotoxin (α7-nicotinic acetylcholine receptor blocker; Sigma-Aldrich) for the recording of electrical excitability. Internal saline consisted of 2 mM MgCl2⋅6H2O, 2 mM EGTA, 5 mM KCl, 20 mM Hepes, and 140 mM K-d-gluconic acid. Previous work has established that the RP2 and aCC motoneurons receive identical synaptic input and can therefore be used interchangeably for the analysis of synaptic input (43). However, these neurons are not identical in their electrical properties; for instance, RP2 displays a characteristic delay to first spike (82). Therefore, in the study of electrical properties we focused on the electrical properties of the aCC motoneuron only.

Immunohistochemistry.

Larvae were staged and nerve cords dissected as described above and fixed in (40mg/ml) formaldehyde (Fisher Scientific) in Sorensen’s saline (pH 7.2, 0.075 M) for 30 min at room temperature. Nerve cords were washed in Sorensen’s containing 0.3% Triton-X-100 (Sigma-Aldrich) and incubated overnight in primary antibodies at 4 °C. After washing they were incubated in secondary antibodies for 2 h at room temperature, mounted in Vectashield supplemented with DAPI (Vector Laboratories), and imaged with a Leica SP5 confocal laser-scanning microscope. For time line experiments, all samples were incubated on the same cover glass to allow for direct comparisons of signal intensities between specimens. Primary antibodies were as follows: EcR-common: mAb Ag10.2, EcR-A: mAb15G1a, and EcR-B1: mAb AD4.4 (Developmental Studies Hybridoma Bank, University of Iowa).

Statistical Analysis.

Quantitative data from dendritic reconstructions were exported from AMIRA as csv files. Data from all experiments were plotted and analyzed using Prism software (GraphPad Software). Student t test, Mann–Whitney U tests and ANOVA with Tukey posttest were used, as appropriate.

Acknowledgments

We thank Michael Bender for generously providing fly stocks. Other fly stocks were obtained from the Bloomington Stock Center. The monoclonal antibodies Ag10.2, 15G1a, and AD4.4, developed by W. S. Talbot and colleagues, were obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the National Institute of Child Health and Human Development and maintained by the University of Iowa, Department of Biological Sciences. We thank Michael Bate, Richard Baines, Jimena Berni, and Darren Williams and three anonymous reviewers for helpful suggestions. We also thank Richard Baines and members of his laboratory at the University of Manchester for training M.F.Z. in electrophysiological techniques, Stefan R. Pulver for help with setting up electrophysiological equipment, and Jason Worrell for preliminary patch-clamp electrophysiology experiments. M.L. was a Royal Society University Research Fellow. The Wellcome Trust supported this work by studentships under the auspices of the four-year Ph.D. program in Developmental Biology to M.F.Z. and O.R., and Wellcome Trust Program Grant WT075934 (to Michael Bate and M.L.). The work benefited from facilities supported by Wellcome Trust Equipment Grant WT079204 and contributions by the Sir Isaac Newton Trust, Cambridge.

Footnotes

  • ↵1Present address: Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138.

  • ↵2J.F.E. and M.L. contributed equally to this work.

  • ↵3To whom correspondence should be addressed. E-mail: jan-felix.evers{at}cos.uni-heidelberg.de.

  • Author contributions: M.F.Z., J.F.E., and M.L. designed research; M.F.Z., O.R., J.F.E., and M.L. performed research; M.F.Z., J.F.E., and M.L. analyzed data; and M.F.Z., J.F.E., and M.L. wrote the paper.

  • The authors declare no conflict of interest.

  • This article is a PNAS Direct Submission.

  • This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1311711110/-/DCSupplemental.

Freely available online through the PNAS open access option.

References

  1. ↵
    1. Thompson DW
    (1992) On Growth and Form (Cambridge Univ Press, Cambridge, UK).
  2. ↵
    1. Chittka L,
    2. Niven J
    (2009) Are bigger brains better? Curr Biol 19(21):R995–R1008.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Purves D,
    2. Lichtman JW
    (1985) Geometrical differences among homologous neurons in mammals. Science 228(4697):298–302.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Snider WD
    (1987) The dendritic complexity and innervation of submandibular neurons in five species of mammals. J Neurosci 7(6):1760–1768.
    OpenUrlAbstract
  5. ↵
    1. Ivanov A,
    2. Purves D
    (1989) Ongoing electrical activity of superior cervical ganglion cells in mammals of different size. J Comp Neurol 284(3):398–404.
    OpenUrlCrossRefPubMed
  6. ↵
    Kaznowski C, Schneiderman H, Bryant P (1985) Cuticle secretion during larval growth in Drosophila melanogaster. J Insect Phys 31(10):801–813.
  7. ↵
    1. Keshishian H,
    2. et al.
    (1993) Cellular mechanisms governing synaptic development in Drosophila melanogaster. J Neurobiol 24(6):757–787.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Schuster CM,
    2. Davis GW,
    3. Fetter RD,
    4. Goodman CS
    (1996) Genetic dissection of structural and functional components of synaptic plasticity. I. Fasciclin II controls synaptic stabilization and growth. Neuron 17(4):641–654.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Kohsaka H,
    2. Okusawa S,
    3. Itakura Y,
    4. Fushiki A,
    5. Nose A
    (2012) Development of larval motor circuits in Drosophila. Dev Growth Differ 54(3):408–419.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Pulver SR,
    2. Bucher D,
    3. Simon DJ,
    4. Marder E
    (2005) Constant amplitude of postsynaptic responses for single presynaptic action potentials but not bursting input during growth of an identified neuromuscular junction in the lobster, Homarus americanus. J Neurobiol 62(1):47–61.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Balice-Gordon RJ,
    2. Breedlove SM,
    3. Bernstein S,
    4. Lichtman JW
    (1990) Neuromuscular junctions shrink and expand as muscle fiber size is manipulated: in vivo observations in the androgen-sensitive bulbocavernosus muscle of mice. J Neurosci 10(8):2660–2671.
    OpenUrlAbstract
  12. ↵
    1. Balice-Gordon RJ,
    2. Lichtman JW
    (1990) In vivo visualization of the growth of pre- and postsynaptic elements of neuromuscular junctions in the mouse. J Neurosci 10(3):894–908.
    OpenUrlAbstract
  13. ↵
    1. Wan HI,
    2. et al.
    (2000) Highwire regulates synaptic growth in Drosophila. Neuron 26(2):313–329.
    OpenUrlCrossRefPubMed
  14. ↵
    1. DiAntonio A,
    2. et al.
    (2001) Ubiquitination-dependent mechanisms regulate synaptic growth and function. Nature 412(6845):449–452.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Aberle H,
    2. et al.
    (2002) wishful thinking encodes a BMP type II receptor that regulates synaptic growth in Drosophila. Neuron 33(4):545–558.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Marqués G,
    2. et al.
    (2002) The Drosophila BMP type II receptor Wishful Thinking regulates neuromuscular synapse morphology and function. Neuron 33(4):529–543.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Sweeney ST,
    2. Davis GW
    (2002) Unrestricted synaptic growth in spinster-a late endosomal protein implicated in TGF-beta-mediated synaptic growth regulation. Neuron 36(3):403–416.
    OpenUrlCrossRefPubMed
  18. ↵
    1. McCabe BD,
    2. et al.
    (2003) The BMP homolog Gbb provides a retrograde signal that regulates synaptic growth at the Drosophila neuromuscular junction. Neuron 39(2):241–254.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Korkut C,
    2. et al.
    (2009) Trans-synaptic transmission of vesicular Wnt signals through Evi/Wntless. Cell 139(2):393–404.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Milton VJ,
    2. et al.
    (2011) Oxidative stress induces overgrowth of the Drosophila neuromuscular junction. Proc Natl Acad Sci USA 108(42):17521–17526.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Guan B,
    2. Hartmann B,
    3. Kho YH,
    4. Gorczyca M,
    5. Budnik V
    (1996) The Drosophila tumor suppressor gene, dlg, is involved in structural plasticity at a glutamatergic synapse. Curr Biol 6(6):695–706.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Rasse TM,
    2. et al.
    (2005) Glutamate receptor dynamics organizing synapse formation in vivo. Nat Neurosci 8(7):898–905.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Li H,
    2. Peng X,
    3. Cooper RL
    (2002) Development of Drosophila larval neuromuscular junctions: Maintaining synaptic strength. Neuroscience 115(2):505–513.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Schubiger M,
    2. Wade AA,
    3. Carney GE,
    4. Truman JW,
    5. Bender M
    (1998) Drosophila EcR-B ecdysone receptor isoforms are required for larval molting and for neuron remodeling during metamorphosis. Development 125(11):2053–2062.
    OpenUrlAbstract
  25. ↵
    1. Cherbas L,
    2. Hu X,
    3. Zhimulev I,
    4. Belyaeva E,
    5. Cherbas P
    (2003) EcR isoforms in Drosophila: Testing tissue-specific requirements by targeted blockade and rescue. Development 130(2):271–284.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Brown HLD,
    2. Cherbas L,
    3. Cherbas P,
    4. Truman JW
    (2006) Use of time-lapse imaging and dominant negative receptors to dissect the steroid receptor control of neuronal remodeling in Drosophila. Development 133(2):275–285.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Riddiford LM,
    2. Cherbas P,
    3. Truman JW
    (2000) Ecdysone receptors and their biological actions. Vitam Horm 60:1–73.
    OpenUrlPubMed
  28. ↵
    1. De Lorme KC,
    2. Schulz KM,
    3. Salas-Ramirez KY,
    4. Sisk CL
    (2012) Pubertal testosterone organizes regional volume and neuronal number within the medial amygdala of adult male Syrian hamsters. Brain Res 1460:33–40.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Cooke BM
    (2006) Steroid-dependent plasticity in the medial amygdala. Neuroscience 138(3):997–1005.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Brenowitz EA
    (2004) Plasticity of the adult avian song control system. Ann N Y Acad Sci 1016:560–585.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Goldstein LA,
    2. Sengelaub DR
    (1994) Differential effects of dihydrotestosterone and estrogen on the development of motoneuron morphology in a sexually dimorphic rat spinal nucleus. J Neurobiol 25(7):878–892.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Goldstein LA,
    2. Sengelaub DR
    (1990) Hormonal control of neuron number in sexually dimorphic spinal nuclei of the rat: IV. Masculinization of the spinal nucleus of the bulbocavernosus with testosterone metabolites. J Neurobiol 21(5):719–730.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Ou Y,
    2. Chwalla B,
    3. Landgraf M,
    4. van Meyel DJ
    (2008) Identification of genes influencing dendrite morphogenesis in developing peripheral sensory and central motor neurons. Neural Dev 3:16.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Schmitt S,
    2. Evers JF,
    3. Duch C,
    4. Scholz M,
    5. Obermayer K
    (2004) New methods for the computer-assisted 3-D reconstruction of neurons from confocal image stacks. Neuroimage 23(4):1283–1298.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Evers JF,
    2. Schmitt S,
    3. Sibila M,
    4. Duch C
    (2005) Progress in functional neuroanatomy: Precise automatic geometric reconstruction of neuronal morphology from confocal image stacks. J Neurophysiol 93(4):2331–2342.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Sholl DA
    (1953) Dendritic organization in the neurons of the visual and motor cortices of the cat. J Anat 87(4):387–406.
    OpenUrlPubMed
  37. ↵
    1. Colombani J,
    2. et al.
    (2005) Antagonistic actions of ecdysone and insulins determine final size in Drosophila. Science 310(5748):667–670.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Riddiford L
    (1993) in The Development of Drosophila Melanogaster, eds Bate CM, Martinez Arias A (Cold Spring Harbor Lab Press, Cold Spring Harbor, NY), pp 889–940.
  39. ↵
    1. Truman JW,
    2. Talbot WS,
    3. Fahrbach SE,
    4. Hogness DS
    (1994) Ecdysone receptor expression in the CNS correlates with stage-specific responses to ecdysteroids during Drosophila and Manduca development. Development 120(1):219–234.
    OpenUrlAbstract
  40. ↵
    1. Talbot WS,
    2. Swyryd EA,
    3. Hogness DS
    (1993) Drosophila tissues with different metamorphic responses to ecdysone express different ecdysone receptor isoforms. Cell 73(7):1323–1337.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Baines RA,
    2. Uhler JP,
    3. Thompson A,
    4. Sweeney ST,
    5. Bate M
    (2001) Altered electrical properties in Drosophila neurons developing without synaptic transmission. J Neurosci 21(5):1523–1531.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Rall W
    (1959) Branching dendritic trees and motoneuron membrane resistivity. Exp Neurol 1:491–527.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Baines RA,
    2. Bate M
    (1998) Electrophysiological development of central neurons in the Drosophila embryo. J Neurosci 18(12):4673–4683.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Burrows M
    (1996) The Neurobiology of an Insect Brain (Oxford Univ Press, Oxford).
  45. ↵
    1. Kuehn C,
    2. Duch C
    (2013) Putative excitatory and putative inhibitory inputs are localised in different dendritic domains in a Drosophila flight motoneuron. Eur J Neurosci 37(6):860–875.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Turrigiano GG,
    2. Leslie KR,
    3. Desai NS,
    4. Rutherford LC,
    5. Nelson SB
    (1998) Activity-dependent scaling of quantal amplitude in neocortical neurons. Nature 391(6670):892–896.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Turrigiano GG
    (2008) The self-tuning neuron: Synaptic scaling of excitatory synapses. Cell 135(3):422–435.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Häusser M,
    2. Spruston N,
    3. Stuart G
    (2008) Dendrites (Oxford Univ Press, Oxford).
  49. ↵
    1. Letinsky MS
    (1974) The development of nerve-muscle junctions in Rana catesbeiana tadpoles. Dev Biol 40(1):129–153.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Davis GW
    (2006) Homeostatic control of neural activity: From phenomenology to molecular design. Annu Rev Neurosci 29:307–323.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Turrigiano G
    (2012) Homeostatic synaptic plasticity: Local and global mechanisms for stabilizing neuronal function. Cold Spring Harb Perspect Biol 4(1):a005736.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Dryer L,
    2. Graziadei PP
    (1994) Mitral cell dendrites: A comparative approach. Anat Embryol (Berl) 189(2):91–106.
    OpenUrlPubMed
  53. ↵
    1. Wittenberg GM,
    2. Wang SS-H
    (2008) in Dendrites, eds Häusser M, Spruston N, Stuart G (Oxford Univ Press, Oxford), pp 43–67.
  54. ↵
    1. Lee T,
    2. Marticke S,
    3. Sung C,
    4. Robinow S,
    5. Luo L
    (2000) Cell-autonomous requirement of the USP/EcR-B ecdysone receptor for mushroom body neuronal remodeling in Drosophila. Neuron 28(3):807–818.
    OpenUrlCrossRefPubMed
  55. ↵
    1. Kuo CT,
    2. Jan LY,
    3. Jan YN
    (2005) Dendrite-specific remodeling of Drosophila sensory neurons requires matrix metalloproteases, ubiquitin-proteasome, and ecdysone signaling. Proc Natl Acad Sci USA 102(42):15230–15235.
    OpenUrlAbstract/FREE Full Text
  56. ↵
    1. Williams DW,
    2. Truman JW
    (2005) Cellular mechanisms of dendrite pruning in Drosophila: Insights from in vivo time-lapse of remodeling dendritic arborizing sensory neurons. Development 132(16):3631–3642.
    OpenUrlAbstract/FREE Full Text
  57. ↵
    1. Dalton JE,
    2. Lebo MS,
    3. Sanders LE,
    4. Sun F,
    5. Arbeitman MN
    (2009) Ecdysone receptor acts in fruitless- expressing neurons to mediate drosophila courtship behaviors. Curr Biol 19(17):1447–1452.
    OpenUrlCrossRefPubMed
  58. ↵
    1. Attwell D,
    2. Laughlin SB
    (2001) An energy budget for signaling in the grey matter of the brain. J Cereb Blood Flow Metab 21(10):1133–1145.
    OpenUrlCrossRefPubMed
  59. ↵
    1. Layalle S,
    2. Arquier N,
    3. Léopold P
    (2008) The TOR pathway couples nutrition and developmental timing in Drosophila. Dev Cell 15(4):568–577.
    OpenUrlCrossRefPubMed
  60. ↵
    1. Voyvodic JT
    (1989) Peripheral target regulation of dendritic geometry in the rat superior cervical ganglion. J Neurosci 9(6):1997–2010.
    OpenUrlAbstract
  61. ↵
    1. Purves D,
    2. Njå A
    (1976) Effect of nerve growth factor on synaptic depression after axotomy. Nature 260(5551):535–536.
    OpenUrlCrossRefPubMed
  62. ↵
    1. Njå A,
    2. Purves D
    (1978) The effects of nerve growth factor and its antiserum on synapses in the superior cervical ganglion of the guinea-pig. J Physiol 277:53–75.
    OpenUrlPubMed
  63. ↵
    1. Zhu B,
    2. et al.
    (2008) Drosophila neurotrophins reveal a common mechanism for nervous system formation. PLoS Biol 6(11):e284.
    OpenUrlCrossRefPubMed
  64. ↵
    1. Fuentes-Medel Y,
    2. et al.
    (2012) Integration of a retrograde signal during synapse formation by glia-secreted TGF-β ligand. Curr Biol 22(19):1831–1838.
    OpenUrlCrossRefPubMed
  65. ↵
    1. Packard M,
    2. et al.
    (2002) The Drosophila Wnt, wingless, provides an essential signal for pre- and postsynaptic differentiation. Cell 111(3):319–330.
    OpenUrlCrossRefPubMed
  66. ↵
    1. Ataman B,
    2. et al.
    (2008) Rapid activity-dependent modifications in synaptic structure and function require bidirectional Wnt signaling. Neuron 57(5):705–718.
    OpenUrlCrossRefPubMed
  67. ↵
    1. Jan LY,
    2. Jan YN
    (1976) Properties of the larval neuromuscular junction in Drosophila melanogaster. J Physiol 262(1):189–214.
    OpenUrlAbstract/FREE Full Text
  68. ↵
    1. Haghighi AP,
    2. et al.
    (2003) Retrograde control of synaptic transmission by postsynaptic CaMKII at the Drosophila neuromuscular junction. Neuron 39(2):255–267.
    OpenUrlCrossRefPubMed
  69. ↵
    1. Davis GW,
    2. Goodman CS
    (1998) Synapse-specific control of synaptic efficacy at the terminals of a single neuron. Nature 392(6671):82–86.
    OpenUrlCrossRefPubMed
  70. ↵
    1. Paradis S,
    2. Sweeney ST,
    3. Davis GW
    (2001) Homeostatic control of presynaptic release is triggered by postsynaptic membrane depolarization. Neuron 30(3):737–749.
    OpenUrlCrossRefPubMed
  71. ↵
    1. Frank CA,
    2. Kennedy MJ,
    3. Goold CP,
    4. Marek KW,
    5. Davis GW
    (2006) Mechanisms underlying the rapid induction and sustained expression of synaptic homeostasis. Neuron 52(4):663–677.
    OpenUrlCrossRefPubMed
  72. ↵
    1. Tripodi M,
    2. Evers JF,
    3. Mauss A,
    4. Bate M,
    5. Landgraf M
    (2008) Structural homeostasis: Compensatory adjustments of dendritic arbor geometry in response to variations of synaptic input. PLoS Biol 6(10):e260.
    OpenUrlCrossRefPubMed
  73. ↵
    1. Jan YN,
    2. Jan LY
    (1978) Genetic dissection of short-term and long-term facilitation at the Drosophila neuromuscular junction. Proc Natl Acad Sci USA 75(1):515–519.
    OpenUrlAbstract/FREE Full Text
  74. ↵
    1. Sen K,
    2. Jorge-Rivera JC,
    3. Marder E,
    4. Abbott LF
    (1996) Decoding synapses. J Neurosci 16(19):6307–6318.
    OpenUrlAbstract/FREE Full Text
  75. ↵
    1. Nicholls JG,
    2. Martin AR,
    3. Fuchs PA,
    4. Brown DA,
    5. Diamond ME
    (2011) From Neuron to Brain (Sinauer Associates, Sunderland, MA).
  76. ↵
    1. Rohrbough J,
    2. Broadie K
    (2002) Electrophysiological analysis of synaptic transmission in central neurons of Drosophila larvae. J Neurophysiol 88(2):847–860.
    OpenUrlAbstract/FREE Full Text
  77. ↵
    1. Henny P,
    2. et al.
    (2012) Structural correlates of heterogeneous in vivo activity of midbrain dopaminergic neurons. Nat Neurosci 15(4):613–619.
    OpenUrlCrossRefPubMed
  78. ↵
    1. Mee CJ,
    2. Pym ECG,
    3. Moffat KG,
    4. Baines RA
    (2004) Regulation of neuronal excitability through pumilio-dependent control of a sodium channel gene. J Neurosci 24(40):8695–8703.
    OpenUrlAbstract/FREE Full Text
  79. ↵
    1. Roy B,
    2. et al.
    (2007) Metamorphosis of an identified serotonergic neuron in the Drosophila olfactory system. Neural Dev 2:20.
    OpenUrlCrossRefPubMed
  80. ↵
    1. Fujioka M,
    2. et al.
    (2003) Even-skipped, acting as a repressor, regulates axonal projections in Drosophila. Development 130(22):5385–5400.
    OpenUrlAbstract/FREE Full Text
  81. ↵
    1. Olver FWJ,
    2. Lozier DW,
    3. Boisvert RF
    (2010) NIST Handbook of Mathematical Functions (Cambridge Univ Press, Cambridge, UK).
  82. ↵
    1. Choi JC,
    2. Park D,
    3. Griffith LC
    (2004) Electrophysiological and morphological characterization of identified motor neurons in the Drosophila third instar larva central nervous system. J Neurophysiol 91(5):2353–2365.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
Bigger dendrites increase activity in growing CNS
Maarten F. Zwart, Owen Randlett, Jan Felix Evers, Matthias Landgraf
Proceedings of the National Academy of Sciences Oct 2013, 110 (40) E3878-E3887; DOI: 10.1073/pnas.1311711110
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 110 (40)
Table of Contents

Submit

Article Classifications

Jump to section

You May Also be Interested in

Greenhouse gas emissions in the first decades of the 21st century are likely to affect global sea level for centuries and limiting long-term sea level rise might require stringent near-term emission reductions. Image courtesy of Lucy Reading-Ikkanda.
Sea-level rise and Paris Agreement pledges
Greenhouse gas emissions in the first decades of the 21st century are likely to affect global sea level for centuries, and limiting long-term sea-level rise might require stringent near-term emission reductions.
Image courtesy of Lucy Reading-Ikkanda (artist).
A single formulation combining two gene therapies treated an array of age-related disorders in mice. Image courtesy of Pixabay /Sipa.
Multidisease gene therapy in mice
A single formulation combining two gene therapies treated an array of age-related disorders in mice.
Image courtesy of Pixabay/Sipa.

Similar Articles