Skip to main content
eLife logoLink to eLife
. 2022 Mar 30;11:e76464. doi: 10.7554/eLife.76464

Steroid hormone signaling activates thermal nociception during Drosophila peripheral nervous system development

Jacob S Jaszczak 1,2, Laura DeVault 1,3, Lily Yeh Jan 1,2, Yuh Nung Jan 1,2,
Editors: Mani Ramaswami4, K VijayRaghavan5
PMCID: PMC8967384  PMID: 35353036

Abstract

Sensory neurons enable animals to detect environmental changes and avoid harm. An intriguing open question concerns how the various attributes of sensory neurons arise in development. Drosophila melanogaster larvae undergo a behavioral transition by robustly activating a thermal nociceptive escape behavior during the second half of larval development (third instar). The Class IV dendritic arborization (C4da) neurons are multimodal sensors which tile the body wall of Drosophila larvae and detect nociceptive temperature, light, and mechanical force. In contrast to the increase in nociceptive behavior in the third instar, we find that ultraviolet light-induced Ca2+ activity in C4da neurons decreases during the same period of larval development. Loss of ecdysone receptor has previously been shown to reduce nociception in third instar larvae. We find that ligand-dependent activation of ecdysone signaling is sufficient to promote nociceptive responses in second instar larvae and suppress expression of subdued (encoding a TMEM16 channel). Reduction of subdued expression in second instar C4da neurons not only increases thermal nociception but also decreases the response to ultraviolet light. Thus, steroid hormone signaling suppresses subdued expression to facilitate the sensory switch of C4da neurons. This regulation of a developmental sensory switch through steroid hormone regulation of channel expression raises the possibility that ion channel homeostasis is a key target for tuning the development of sensory modalities.

Research organism: D. melanogaster

eLife digest

During their lives, animals encounter a broad range of stimuli from their surroundings including heat, light and touch. The ability to appropriately respond to such stimuli is crucial for survival as it allows the animals to avoid predators and other dangers, locate food and shelter, and find mates.

Fruit fly larvae are a useful model for studying how animals respond to unpleasant (known as painful) heat stimuli. When something hot touches a larva, the larva rolls away to avoid the stimulus. The heat stimulates electrical activity in a type of neuron known as C4da neurons on the surface of the larva. Ultraviolet light and several other stimuli are also able to activate electrical activity in C4da neurons, resulting in the larvae changing the direction they move to avoid the stimuli.

Only older fly larvae respond to painful heat stimuli and previous studies found that a hormone receptor protein is required for this response. However, it remains unclear how this response develops as the larvae age.

Jaszczak et al. studied the behavior of fly larvae and electrical activities of C4da neurons in response to painful heat and ultraviolet light. The experiments found that painful heat triggered more rolling behavior from older larvae than those of younger larvae. In contrast, ultraviolet light triggered lower levels of electrical activity in the C4da neurons of older larvae than those of younger larvae. The team raised the levels of a hormone known as ecdysone and found that this increased the rolling behavior in younger larvae. They then increased the amount of receptor protein for this hormone in the neurons and found that it decreased the levels of another protein called Subdued in the C4da neurons. This in turn increased the neurons’ response to painful heat and decreased their response to ultraviolet light.

Jaszczak et al. propose that as the larva develops, ecdysone reduces the levels of Subdued, which promotes C4da neurons to switch their sensitivity from detecting ultraviolet light to painful heat. In the future, better understanding of what causes pain sensations in developing animals will help us search for factors that cause long-term pain conditions in humans.

Introduction

Detection of internal and external stimuli depends on developmental programing of the proper physiological and morphological properties of sensory neurons. It is an intriguing open question as to how the characteristics of sensory transduction arise during development, and which neuronal properties are crucial for the initiation of sensory detection.

Nociception, the ability to detect and escape potentially harmful stimuli, is a highly conserved function of sensory systems present in all animals (Arenas et al., 2017). A nociception system is composed of nociceptors, which are neurons with molecular components that directly sense harmful stimuli; a processing center where the stimuli are interpreted, typically the neural circuits of the central nervous system (CNS); and the reflexive neuromuscular circuit which generates a reaction to escape injury (Baliki and Apkarian, 2015). Development of sensory systems requires the coordination of processes on the cellular, morphogenic, and circuit level.

Thermal nociception generates an avoidance behavior, such as a fast limb retraction or change of movement trajectory. Previous studies of thermal nociception behaviors and neural activity demonstrate a change in nociceptive responses during development of both mice and Drosophila. For example, a mouse paw exhibits low sensitivity to painful heat early in postnatal development, but the rate of reflex gets faster as the animal develops (Jankowski et al., 2014; Ford et al., 2019). A developmental transition is also observed in nociceptor activity, as revealed by the change in the number of heat-sensitive C-fibers with age in mice (Hiura et al., 1992). Likewise, a behavioral transition in heat thermal nociception has been observed in Drosophila melanogaster larvae (Sulkowski et al., 2011). Drosophila third instar larvae exhibit a stereotyped ‘corkscrew’ behavior, when a thermal probe heated above 38°C touches the cuticle (Tracey et al., 2003; Babcock et al., 2009). However, this behavior is not observed in second instar or earlier larval stages, and appears to be acquired only during the last stage of larval development.

The larval nociceptive behavior has been a useful model for dissection of the molecular mechanisms and circuitry of nociception (Himmel et al., 2017). Among the heat-sensitive neurons in the larval cuticle (Liu et al., 2003) are the Class IV dendritic arborization (C4da) neurons, which function as the primary nociceptors for noxious heat-induced escape behavior (Tracey et al., 2003; Hwang et al., 2007). C4da neurons are multimodal sensors for harmful stimuli, with the capacity to respond to multiple forms of sensory information. In addition to responding to temperature, C4da neurons respond to high-intensity blue light (Xiang et al., 2010), noxious UV light (Xiang et al., 2010; Gu et al., 2019), and harsh touch (Zhong et al., 2010).

While noxious heat and touch both elicit the rolling escape behavior, stimulation of C4da neurons with short wavelength (blue or UV) light causes a locomotion trajectory avoidance behavior. These different stimuli cause different electrical signals in C4da neurons. Noxious heat causes high-frequency spike trains interspersed with quiescent periods, in contrast to blue light which elicits lower frequency firing with few quiescent periods in third instar larvae (Terada et al., 2016). Therefore, distinct behaviors appear to be encoded by the C4da neurons according to their differential responses to differing stimuli.

The majority of studies of nociception in Drosophila larvae have focused on the last (i.e., the third) larval instar, where nearly all larvae will perform the behavioral response. However, Sulkowski et al. found that during the first half of larval development (i.e., the first and second instars), larvae exhibited either no or a low rate of thermal nociceptive behavior, leading to the hypothesis that a developmental transition occurs between the second and third larval instars (Sulkowski et al., 2011). The mechanisms which regulate this transition remain unknown. Optogenetic activation of C4da neurons can induce the corkscrew behavior or a locomotion trajectory avoidance behavior in third instar larvae. However, optogenetic activation could only induce a locomotion trajectory avoidance behavior in the second instar. This difference between second and third instars in the optogenetic activation of behavior, taken together with the difference in coding activity in C4da neurons in response to thermal stimulation or light (Terada et al., 2016), raises the possibility that the developmental transition may involve a change in sensory coding. Supporting this possibility, misexpression of low temperature activated TrpA1 (isoform A) channels in the C4da neurons enables second instar larvae to exhibit corkscrew behavior in response to an innocuous temperature stimulus (Luo et al., 2017). It is conceivable that the sensory switch may be caused by a cell autonomous change in the C4da neurons. Whether C4da neurons change in their neural activity during the developmental transition from second to third instar is unknown, and the regulators of such transition are unknown as well.

Transcription factors facilitate the formation of distinct neuronal functions (Parrish et al., 2014) and regulate the dynamic expression profiles of many developmental transitions between and during larval instars (Sullivan and Thummel, 2003). One such transcription factor is the nuclear hormone receptor ecdysone receptor (EcR). EcR forms a heterodimer with ultraspiracle (USP) and binds the ligand ecdysone, a steroid hormone, to coordinate a wide variety of developmental programs (Yamanaka et al., 2013). EcR and USP are required for thermal nociception in the third instar (McParland et al., 2015), but the mechanism and role of ecdysone signaling in the second instar have not been determined.

EcR expression in C4da neurons has been reported at the beginning and end of larval development (Kuo et al., 2005; Ou et al., 2008; Kirilly et al., 2009; McParland et al., 2015). In the absence of ligand, EcR and USP distribute between the cytoplasm and nucleus, while ligand binding increases nuclear localization (Nieva et al., 2007; Nieva et al., 2008). Without ligand binding, EcR and USP recruit nucleosome remodeling proteins to compact the chromatin and locally suppress transcription. When bound to ecdysone, EcR-USP recruit coactivators, such as histone acetylases, to expand the chromatin structure and promote transcription. In this way, steroid hormone signaling can create positive and negative feedback to promote cascades of transcriptional programs at developmentally precise periods (Sullivan and Thummel, 2003; Hill et al., 2013). Whether the ligand-induced activity and transcriptional regulation are involved in EcR-mediated thermal nociception remains to be determined.

In this study, we performed a series of experiments to determine the activity of the C4da neurons during the developmental nociceptive transition as well as the mechanisms which regulate this sensory switch. These experiments demonstrate that steroid hormone signaling in nociceptive neurons regulates the channel composition in C4da neurons to modulate the sensory properties and behavioral response at distinct stages of larval development.

Results

C4da neuron sensory switch between second and third larval instars is modality specific

C4da neurons are multimodal nociceptive sensors of thermal, short wavelength light, and mechanical stimuli (Himmel et al., 2017). While first instar and third instar larvae respond at similar frequency to mechanonociceptive stimuli (Almeida-Carvalho et al., 2017), larvae do not robustly activate the thermally triggered nociceptive behavior until the last (third) instar of development (Sulkowski et al., 2011). Whether the C4da neurons respond to short wavelength light changes during larval development is unknown. To investigate whether this developmental transition is specific to thermal nociception, we sought to compare the responses of second and third instar larvae to thermal and light stimuli.

First, we measured the frequency and latency of the nociceptive behavior in second and third instar larvae across a range of nociceptive temperatures. Compared to third instar larvae, second instar larvae had a lower response frequency of nociceptive behavior across all nociceptive temperatures above 40°C (Figure 1A). Among the larvae which did display a nociceptive behavior, second instar larvae had a longer latency before initiating the nociceptive response than third instar larvae (Figure 1B, Figure 1—figure supplement 1A, B). Thus, second instar larvae displayed reduced thermal nociceptive behavior across a wide range of temperatures.

Figure 1. Thermal nociception behaviors increase during larval development, while UV-A light response decreases.

(A) Groups of age-matched control larvae were touched with a thermal probe at different temperatures. The nociceptive response was scored if a 360° roll occurred during 20 s of stimulus. (B) Latency by which 50% of the responding population of larvae has had a nociceptive behavior response. (C) Representative GCaMP6s images of second and third instars in soma Class IV dendritic arborization (C4da) neurons. Scale bar = 10 μm. Periods of UV-A treatment program: Pre = before UV-A stimulus. During = period of UV-A light stimulus. Post = after the end of UV-A stimulus. (D) Individual traces (gray lines) and means (red lines) of Ca2+ activity calculated by ratiometric change from baseline. Gray column indicates period of UV-A stimulus. Dotted line indicates threshold level for % over threshold calculations. (E) Percent of soma with Ca2+ activity over the threshold level. (F) Peak Ca2+ activity during the periods of UV-A treatment programs. (G) Heat and light development have opposite trends during larval development. (A) n = 2–4 staging replicates of 15–20 larvae were tested for each age and temperature. (C–F) Second instar n = 21 neurons. Third instar n = 19 neurons. (A, B) Student’s t-test. (E) Fisher’s exact test, one-sided. (F) Student’s t-test. *p < 0.05, **p < 0.01.

Figure 1.

Figure 1—figure supplement 1. Thermal nociceptive sensitivity increases across temperatures from second to third instar.

Figure 1—figure supplement 1.

Accumulated curve of percentage larvae that will have exhibited the nociceptive behavior by the length of time experiencing the stimulus (latency in seconds). (A) Percentage of second instar larvae that will have had the nociceptive behavior at the latency response time. (B) Percentage of third instar larvae that will have had the nociceptive behavior at the latency response time. Quantification of data from Figure 1A. n = 2–4 staging replicates of 15–20 animals were tested for each age and temperature.

High-intensity UV light exposure of third instar larvae causes behavioral responses, increased Ca2+ activity, and patterns of spike trains that are different from those caused by nociceptive temperatures (Xiang et al., 2010; Terada et al., 2016). The small size of second instar larvae makes electrophysiological recording of C4da neurons prohibitive, therefore, we sought to compare the magnitude of the UV-induced Ca2+ response. To determine if the UV light-induced Ca2+ activity response of C4da neurons changes during development, we measured GCaMP6s signals of these sensory neurons in second and third instar larvae in response to 405 nm (UV-A) light stimulation. In contrast to the transition of thermal nociceptive behavior, UV-A stimuli-induced Ca2+ activity of C4da neurons in second instar larvae was higher than that in third instar larvae (Figure 1C, D). This pattern, of third instar C4da neurons having lower UV-A-induced Ca2+ activity than second instar C4da neurons, was validated by quantifying both the number of neurons which surpassed a threshold value of UV-A-induced Ca2+ activity and the magnitude of the peak Ca2+ activity (Figure 1E, F). Thus, the UV-A-induced Ca2+ activity decreased with larval development, in contrast to the increase of thermal nociceptive neuronal response from second to third instar (Figure 1G). Based on these observations, we conclude that regulation of the sensory switch during the second and third instar development is modality specific.

Steroid hormone ecdysone promotes the development of thermal nociception

Having found modality-specific sensory responses during the second to third instar developmental transition, we sought to find regulators of the sensory switch in C4da neurons by first examining the thermal nociceptive transition. The steroid hormone ecdysone regulates multiple developmental events during the second and third instars, with peaks in ecdysone titer triggering molting events and developmental checkpoints (Rewitz et al., 2013). Ecdysone is the ligand for the EcR with three isoforms (A, B1, and B2). These isoforms are identical in the DNA- and ligand-binding domains but differ in the activation function 1 (AF1) domain (Figure 2—figure supplement 1A). C4da neurons express isoforms EcR-A and EcR-B1 at embryonic stages, during the third instar, and dynamically during pupal development (Kuo et al., 2005; Ou et al., 2008; Kirilly et al., 2009; McParland et al., 2015). EcR-B1 regulates dendrite growth (Ou et al., 2008) and pruning (Kuo et al., 2005; Kirilly et al., 2009). EcR-A is required for thermal nociception in the third instar, as revealed by the greater reduction of nociception caused by RNAi knockdown of EcR-A as compared to RNAi knockdown of EcR-B1 (McParland et al., 2015). Therefore, we investigated the role of ecdysone and EcR-A in the developmental transition of thermal nociceptive behavior.

Nuclear localization of EcR increases during the end of larval development; localization is greater in late third instar larvae and pupae as compared to early third instar larvae (Kuo et al., 2005; Kirilly et al., 2009). However, the expression and localization of EcR isoforms during the second larval instars are unknown. To investigate EcR dynamics during the developmental period with different levels of thermal nociception, we examined the localization of EcR in second and third instar larval C4da neurons with an antibody which recognizes a common domain of all EcR isoforms. Consistent with previous findings (Kuo et al., 2005; Ou et al., 2008; Kirilly et al., 2009; McParland et al., 2015), we observed nuclear localization of EcR in wandering third instar larvae (Figure 2A, B). Earlier in larval development, EcR was evenly distributed between the nucleus and cytoplasm in the second instar (2 and 8 hr After L2 Ecdysis) and at the beginning of the third instar, 2 hr After L3 Ecdysis (AL3E). Nuclear localization increased 8 hr AL3E and in wandering third instar larvae (Figure 2A, B). Similar EcR localization dynamics could be detected for specific EcR isoforms, as revealed by antibodies specific to EcR-A (Figure 2C, D) and EcR-B1 (Figure 2E, F). Quantification of total EcR immunoreactivity in C4da neurons did not reveal any significant change during the second and third instar development (Figure 2—figure supplement 1B). These observations suggest that while EcR is present in C4da neurons throughout the second and third instars, the nuclear localization begins to increase early in the third instar.

Figure 2. Ecdysone receptor (EcR) nuclear localization in Class IV dendritic arborization (C4da) neurons increases early during the third instar.

(A) EcR immunohistochemistry with EcR-common antibody (C) EcR-A antibody, and (E) EcR-B1 antibody. Red dashed line outlines C4da neurons and yellow dashed line indicates position of nucleus. C4da neurons are labeled by ppktd-GFP. EcR nuclear localization quantified by nuclear to cytoplasmic ratio of intensity with (B) EcR-common and (D) EcR-A, and (E) EcR-B1 antibody. After L2 Ecdysis (AL2E), After L3 Ecdysis (AL3E). One-way analysis of variance (ANOVA) with Bonferroni post test. *p < 0.05, **p < 0.01, ****p < 0.0001.

Figure 2.

Figure 2—figure supplement 1. Domains of ecdysone receptor (EcR) isoforms and total EcR in Class IV dendritic arborization (C4da) neuron early during the second and third instars.

Figure 2—figure supplement 1.

(A) EcR domain structures. Activation function (AF) and nuclear receptor (NR) domains along with DNA (dark bar) and ligand- (light bar) binding domains. All EcR isoforms have identical sequences except for the A/B NR domain which contains AF1 function. EcR-ΔC is a synthetic isoform containing only sequences present in all isoforms. Scale in amino acid residues. (B) EcR immunohistochemistry with EcR-common antibody quantifying total intensity in both the nucleus and cytoplasm. One-way analysis of variance (ANOVA) with Bonferroni post test.

Changes in ecdysone synthesis and systemic ecdysone titer orchestrate critical periods of Drosophila development. Pulses of ecdysone in the first and second larval instars increase systemic titers and promote expression of genes required for molting, while a series of three pulses of ecdysone in the third instar control developmental checkpoints and developmental preparation for metamorphosis (Warren et al., 2006; Kannangara et al., 2021). To determine whether the increased thermal nociception in third instar larvae is driven by systemic ecdysone, we tested whether increasing steroid hormone titer during the second instar is sufficient to cause precocious development of thermal nociceptive behavior. Feeding larvae food supplemented with the metabolically active 20-hydroxyecdysone (20E) increases ecdysone signaling in larvae (Colombani et al., 2005). Thus, we transferred larvae to 20E supplemented food and measured larval thermal nociceptive behavior 8 hr later, while the larvae were still in the second instar. Larvae fed with 250 μg/ml 20E had a higher rate of behavioral response to thermal nociceptive stimuli as compared to larvae fed with 125 μg/ml 20E or vehicle alone (Figure 3A), indicating that increasing ecdysone titer in the second instar is sufficient to promote the development of thermal nociception.

Figure 3. The nociceptive transition is ecdysone ligand activity dependent.

Figure 3.

(A) Percent of second instar larvae population displaying nociceptive behavior when fed 20-hydroxyecdysone (20E). (B, D) Percent of population displaying nociceptive behavior when overexpressing ligand-binding mutant ecdysone receptor (EcR)-A (ppk-Gal4/UAS-EcR-A -W650A), or coactivator mutant EcR-A (ppk-Gal4/UAS-EcR-A-F645A). Third instar larvae with (B) 42°C nociceptive probe or (D) 46°C nociceptive probe. (C, E) Latency by which 50% of the responding population displaying nociceptive behavior when overexpressing mutant EcR-A. Third instar larvae with (C) 42°C nociceptive probe or (E) 46°C nociceptive probe. One-way analysis of variance (ANOVA) with Tukey’s post test. *, a, or b = p < 0.05. (A) n > 35 larvae for each treatment. (B) n = 2–4 staging replicates of 15–20 larvae were tested for each genotype.

The binding of ecdysone to EcR causes the recruitment of coactivators and the loss of corepressors. In the third instar, reduction of EcR-A expression reduces thermal nociception (McParland et al., 2015). Therefore, we tested the effects of disrupting the EcR-A domains responsible for either ligand binding or coactivator recruitment on nociception, by characterizing mutants with alanine substitution in either the ligand-binding domain (EcR-A-W650A) or the coactivator recruitment domain (EcR-A-F645A) (Cherbas et al., 2003). The W650A mutation prevents ligand binding and disrupts both derepression and activation. The F645A mutation cannot mediate activation, but retains the ligand-binding capacity (Hu et al., 2003). Therefore, differences between the effect of overexpression of these mutant EcR-A constructs are likely due to differences in ligand-binding ability. Mutant forms of EcR can function in a dominant negative capacity by outcompeting the endogenous wild-type EcR, thereby inactivating the endogenous receptor function (Cherbas et al., 2003). In this way, overexpression of these mutant EcR-A constructs can distinguish the requirements of coactivation (disrupted by F645A) or ligand-induced derepression (disrupted by W650A). When assaying for a dominant negative mutant phenotype, we observed that expression of EcR-A-F645A in C4da neurons did not change nociceptive behavior in third instar larvae, while expression of EcR-A-W650A reduced the number of larvae exhibiting nociceptive behavior in third instar larvae (Figure 3B). Of the larvae that did exhibit nociceptive behavior, expression of EcR-A-W650A did not alter their response latency relative to the W650A controls, but did increase the latency relative to F645A genotypes (Figure 3C). We conclude that ecdysone ligand-binding activity of EcR-A is required for nociception in third instar larvae.

TrpA1 knockout larvae lacking expression of all TrpA1 isoforms lose nociceptive behavior in response to thermal stimuli at 40–44°C, but are still able to respond to a 46°C probe with nociceptive behavior (Gu et al., 2019). Therefore, we sought to determine whether disruption of EcR function has effects specific to the probe temperature. Interestingly, EcR-A-W650A expression affected nociceptive behavior induced with the 42°C probe (Figure 3B, C) but not with the 46°C probe (Figure 3D, E). It thus appears that EcR-A coactivator assembly is not required for third instar nociception, while ligand-binding activity is required for third instar nociceptive behavior in response to a 42°C probe.

Having found that ecdysone ligand activity through EcR-A is required for nociception, we sought to determine if overexpression of EcR-A in C4da neurons is sufficient to render second instar larvae precociously nociceptive. Since second instar larvae have a low thermal nociceptive response rate, we assayed for a gain of thermal nociceptive response in second instar larvae with EcR-A overexpression in their C4da neurons. Indeed, overexpression of EcR-A was sufficient to increase thermal nociceptive behavior in second instar larvae (Figure 4A, C). In contrast to the temperature specific effect of dominant negative mutant expression in the third instar, overexpression of EcR-A in the second instar was able to promote nociception at both 42 and 46°C probe temperatures (Figure 4A, C, E). Of the larvae which did display a nociceptive behavior, overexpression of EcR-A did not significantly change the latency of the nociceptive response as compared to second instar control larvae (Figure 4B).

Figure 4. Ecdysone receptor (EcR)-A overexpression in Class IV dendritic arborization (C4da) neurons increases thermal nociception in the second instar.

(A) Nociceptive behavior in second instar larvae with EcR isoforms overexpressed in C4da neurons. (C, E) Nociceptive behavior in second instar larvae with overexpression of EcR-A mutants. Ligand-binding mutant EcR-A (ppk-Gal4/UAS-EcR-W650A), or coactivator mutant EcR-A (ppk-Gal4/UAS-EcR-F645A). (B, D, F) Second instar latency by which 50% of the responding population displaying nociceptive behavior when overexpressing wild-type or mutant EcR-A. (A–D) 42°C nociceptive probe and (E, F) 46°C nociceptive probe. One-way analysis of variance (ANOVA) with Tukey’s post test. *p < 0.05, **p < 0.01. n = 2–4 staging replicates of 15–20 larvae were tested for each genotype.

Figure 4.

Figure 4—figure supplement 1. Ecdysone receptor (EcR)-A overexpression in Class IV dendritic arborization (C4da) neurons reduces dendrite tip number in the second instar.

Figure 4—figure supplement 1.

(A–D) Second instar C4da neuron arbors overexpressing wild-type EcR-A. (A) Representative images of second instar C4da neuron arbors. (B) Second instar C4da neuron arbor area. (C) Second instar C4da neuron dendrite tip number. (D) Second instar C4da neuron dendrite density. (E, F) Student’s t-test. *p < 0.05. n = 2 staging replicates of 30 and 28 larvae were tested for each genotype, respectively.
Figure 4—figure supplement 2. Ecdysone receptor (EcR) isoform overexpression in Class IV dendritic arborization (C4da) neurons does not change thermal nociception in the third instar.

Figure 4—figure supplement 2.

(A) Nociceptive behavior in third instar larvae with EcR isoforms overexpressed in C4da neurons. (B) Third instar latency by which 50% of the responding population displaying nociceptive behavior when overexpressing EcR isoforms. 42°C nociceptive probe. One-way analysis of variance (ANOVA) with Tukey’s. n = 3 staging replicates of 15–20 larvae were tested for each genotype.

Previous studies have found that overexpression of each EcR isoform can suppress expression of the other EcR isoforms (Schubiger et al., 2003). Additionally, overexpression of each EcR isoform could rescue loss of EcR function mutations in specific tissues (Cherbas et al., 2003). The difference between EcR isoforms is within the A/B domains (Figure 2—figure supplement 1A). Therefore, we tested whether nociception in second instar larvae could be increased by overexpression in C4da neurons of EcR-A, EcR-B1, or EcR-ΔC. EcR-ΔC is an artificial isoform which contains only the sequences common to all isoforms and no A/B domain. We found that neither overexpression of EcR-B1 nor overexpression of EcR-ΔC could increase nociception of second instar larvae (Figure 4A). Additionally, overexpression of these EcR isoforms did not alter nociception in the third instar (Figure 4—figure supplement 2A, B). Given that expression of neither the domains common to all EcR isoforms, which are present in EcR-ΔC, nor the AF1 domain of EcR-B1 could increase nociception in the second instar, we conclude that the A/B domain unique to EcR-A is required to increase nociception of second instar larvae.

Expression of EcR-A RNAi or EcR-B1-dominant negative mutant in the third instar larvae reduces the number of dendrite tips (Ou et al., 2008; McParland et al., 2015). Insensitivity to nociception is often associated with a reduction in dendrite structure, while hypersensitivity to nociception can be associated with both increased and decreased dendrite coverage (Honjo et al., 2016). We examined whether the EcR-A overexpression induced nociception was associated with changes of the dendrite branch number or arbor size. We found that EcR-A overexpression in the second instar reduced the number of dendrite tips (a measure of branch number) without significantly changing the area of the dendrite arbor (Figure 4—figure supplement 1A–D). Our observation of decreased dendrite tip number with an associated increase in nociception is consistent with previously observed trends of decreased dendrite coverage in other nociceptive hypersensitive phenotypes (Honjo et al., 2016), further adding to the evidence that dendrite structure alone cannot predict nociceptive sensitivity.

Having found that ecdysone binding to EcR-A is required for third instar nociception, we sought to determine whether ligand binding is required for the precociously induced second instar nociception. We found that with overexpression of EcR-A-F645A, larvae were still precociously nociceptive as second instar larvae and they responded at a similar level as those overexpressing wild-type EcR-A (Figure 4C), suggesting that coactivator recruitment is not required for EcR-A induction of thermal nociception. In contrast, overexpression of the ligand-binding mutant EcR-A-W650A did not facilitate precocious development of nociceptive behavior in second instar larvae to the level exhibited by larvae overexpressing wildtype EcR-A (Figure 4C). These effects of EcR mutant expression were observed at both 42 and 46°C probe temperatures (Figure 4C, E). Of those larvae which did display a nociceptive behavior, overexpression of mutant EcR-A did not significantly change the latency of the nociceptive response (Figure 4D, F). These findings reveal that thermal nociceptive development in the second instar larvae is sensitive to ecdysone regulation mediated by EcR-A via its ligand-binding domain.

Together, these results demonstrate that EcR-A signaling is necessary and sufficient for the development of thermal nociceptor behavior. EcR-A acts in a cell autonomous manner in the C4da sensory neurons and requires the function of the ligand-binding domain. Our results also suggest that coactivator recruitment is not required for thermal nociceptive development or maintenance. Additionally, our results suggest that in the third instar, the maintenance of ability to detect 46°C is independent of EcR-A ligand binding.

EcR regulates subdued expression during larval thermal nociceptive development

Having found that the EcR-A A/B domain and ligand binding are required for development of thermal nociception, we sought to determine whether any of the following nociceptive sensors are transcriptionally regulated by EcR-A. TrpA1 and Painless are transient receptor potential (TRP) channels required for thermal nociceptive behavior and nociceptive temperature-induced Ca2+ activity in C4da neurons (Tracey et al., 2003; Sokabe et al., 2008; Zhong et al., 2012; Terada et al., 2016; Gu et al., 2019). The TMEM16 family member Subdued is also required for thermal nociceptive behavior (Jang et al., 2015). To assess the scope of EcR-A regulation, we also included the degenerin/epithelial sodium channel (DEG/ENaC) channel Ppk1 that is required for ambient temperature-driven behavior (Ainsley et al., 2008) and also functions along with Ppk26 in mechanical nociception (Gorczyca et al., 2014; Mauthner et al., 2014), as well as Piezo, which is involved in mechanical nociception but not thermal sensation in C4da neurons (Kim et al., 2012).

In order to compare the changes in gene expression during the sensory switch in larval development, we conducted quantitative real-time PCR (qRT-PCR) with the nociceptive genes in GFP-labeled C4da neurons isolated by fluorescence-activated cell sorting (FACS) from control second and third instar larvae. We found decreased expression of TrpA1, subdued, ppk1, and ppk26 in third instar when compared to second instar C4da neurons. In contrast, painless expression increased while piezo expression did not significantly change from second to third instar C4da neurons (Figure 5A).

Figure 5. Ecdysone receptor (EcR) transcriptionally regulates subdued during the period of thermal nociceptive development.

Expression of nociceptive genes measured by qRT-PCR from fluorescence-activated cell sorting (FACS) purified Class IV dendritic arborization (C4da) neurons. (A) C4da neurons isolated from second and third instar larvae. Expression in third instar neurons was normalized to second instar expression. (B) C4da neurons isolated from second instar larvae expressing wild-type EcR-A. Genotypic controls not expressing EcR-A (ppkCD4/UAS-EcR-A) were normalized to age-matched neurons (ppkCD4/+). Expression in EcR-A expressing neurons (ppk-Gal4/UAS-EcR-A) was normalized to age-matched control neurons without EcR-A expression (ppk-Gal4/+). Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001. (A) n = 8 and (B) n = 3–4 FACS isolation/staging replicates for each age and genotype.

Figure 5.

Figure 5—figure supplement 1. Mutant ecdysone receptor (EcR) transcriptionally regulates nociceptive genes during the period of thermal nociceptive development.

Figure 5—figure supplement 1.

Expression of nociceptive genes measured by qRT-PCR from fluorescence-activated cell sorting (FACS) purified Class IV dendritic arborization (C4da) neurons. (A) Second instar and (B) third instar C4da neurons expressing ligand-binding mutant EcR-A-W650A. Genotypic controls not expressing EcR-A-W650A (ppkCD4/UAS- EcR-A-W650A) were normalized to age-matched neurons (ppkCD4/+). Expression in EcR-A-W650A expressing neurons (ppk-Gal4/UAS-EcR-A-W650A) was normalized to control neurons without EcR-A-W650A expression (ppk-Gal4/+). Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001. n = 3–4 FACS isolation/staging replicates for each genotype.
Figure 5—figure supplement 2. Ecdysone receptor A (EcR-A) RNAi reduces EcR-A protein but does not alter nociceptive gene expression.

Figure 5—figure supplement 2.

(A) EcR-A RNAi reduces EcR-A protein expression and does not reduce EcR-B1 protein expression. EcR immunohistochemistry with EcR-B1 or EcRA antibody. Yellow dashed line indicates position of nucleus. Class IV dendritic arborization (C4da) neurons labeled by ppktd-GFP. (B) C4da neurons isolated from third instar larvae expressing EcR-A RNAi. Genotypic controls not expressing EcR-A RNAi (+/EcR-A RNAi) and EcR-A RNAi expressing neurons (ppk >EcR A RNAi) were normalized to age-matched control neurons without EcR-A RNAi expression (ppk-Gal4/+). Transcripts of piezo were not measurable from samples (N/A: not applicable). Student’s t-test. n = 3–4 fluorescence-activated cell sorting (FACS) isolation/staging replicates for each genotype.

Next, we overexpressed EcR-A in C4da neurons and isolated these neurons with FACS from second instar larvae to measure expression of the nociceptive sensors. Comparing transcript levels of EcR-A-overexpressing C4da neurons to controls revealed that subdued expression was suppressed in EcR-A expressing C4da neurons as compared to that in control second instar C4da neurons (Figure 5B), thus mimicking the decrease during developmental transition from second to third instar. Notably, subdued stood out among the channel genes tested as the only one that displayed EcR-A regulation consistent with the developmental sensory switch.

While overexpression of EcR-A increased nociception in second instar larvae, overexpression of the ligand-binding mutant EcR-A-W650A did not increase nociception in second instar but reduced nociception in the third instar (Figures 3 and 4). Having found that EcR-A overexpression specifically suppressed subdued expression, we sought to determine whether expression of the EcR-A-dominant negative mutant has any effect on the level of subdued expression. When we overexpressed EcR-A-W650A in C4da neurons and then used FACS to isolate these neurons from either second or third instar larvae, we observed a reduction of subdued expression at both stages of larval development (Figure 5—figure supplement 1A, B). Additionally, we found that expression EcR-A-W650A in second instar larvae significantly reduced TrpA1, ppk26, and piezo expression (Figure 5—figure supplement 1A), while EcR-A-W650A expression in the third instar larvae significantly reduced ppk1, ppk26, and piezo expression (Figure 5—figure supplement 1B).

Because EcR-A RNAi expression in C4da neurons has been shown to reduce nociception in the third instar, we also measured expression of the nociceptive sensors from FACS isolated third instar C4da neurons which expressed EcR-A RNAi. Expression of the EcR-A RNAi robustly reduced the presence of EcR-A protein in C4da neurons (Figure 5—figure supplement 2A), but we found no significant difference in the expression of nociceptive sensor genes (Figure 5—figure supplement 2B).

These data demonstrate that EcR-A regulates subdued expression. EcR-A overexpression can specifically suppress subdued expression in second instar larvae, as expected from the decrease of subdued expression in third instar larvae. We also found that inhibiting EcR-A ligand-binding capacity with expression of EcR-A-W650A can reduce expression of more nociceptive genes than expression of wild-type EcR-A. EcR-A-W650A expression reduced subdued expression in the second instar, and further reduced expression of subdued in the third instar larvae. We conclude that during the developmental period of the sensory switch, EcR-A regulates expression of subdued, and that ligand-binding activity is necessary for maintaining the regulated expression of multiple nociceptor genes in the third instar.

Reduction of subdued confers second instar nociceptors with thermal and light sensitivity characteristic of third instar nociceptors

Our transcriptional data indicate that EcR-A overexpression specifically suppresses subdued expression in second instar C4da neurons. Therefore, we sought to determine whether subdued expression in second instar C4da neurons is responsible for the suppression of thermal nociceptive behavior in early larval development.

A knockout mutant of the subdued gene locus (subduedKO11) (Wong et al., 2013) eliminated subdued mRNA expression (Figure 6—figure supplement 1A), but did not exhibit a significant change of nociceptive behavior in second instar larvae (Figure 6A). However, subduedKO11 larvae were significantly smaller than age-matched second instar control larvae (Figure 6—figure supplement 1B), raising the possibility that pleiotropic effects of subdued loss of function mutation in the whole animal may alter the behavioral dynamics and mask the effect of mutation in the C4da neurons. Expression of a subdued-RNAi was effective at reducing subdued expression but did not eliminate all expression (Figure 6—figure supplement 1C). Therefore, we tested the effect of expression of subdued-RNAi in C4da neurons, which did not reduce the size of second instar larvae (Figure 6—figure supplement 1D). When subdued-RNAi was expressed specifically in the C4da neurons, second instar larvae had a greater nociceptive behavioral response than controls (Figure 6B). Thus, RNAi reduction of subdued expression in the C4da neurons caused a precocious thermal nociceptive response in second instar larvae, phenocopying EcR-A overexpression.

Figure 6. Subdued-RNAi changes second instar Class IV dendritic arborization (C4da) neuron sensory response pattern to third instar sensory response pattern.

(A) Second instar larvae with 42°C nociceptive probe. Percent of population displaying nociceptive behavior when mutant for Subdued. (B) Second instar larvae with 42°C nociceptive probe. Percent of population displaying nociceptive behavior when expressing Subdued-RNAi (ppk-Gal4 > Subd RNAi v37472). (C) Individual traces (gray lines) and means (red lines) of Ca2+ activity calculated by ratiometric change from baseline. Gray column indicates period of stimulus. Dotted line indicates threshold level for % over threshold calculations. (D) Percent of soma with Ca2+ activity over the threshold level. (E) Peak Ca2+ activity during the periods of stimulus programs. (A, B) One-way analysis of variance (ANOVA). (D) Fisher’s exact test, one-sided. (E) Student’s t-test. (A, B) n = 3–4 staging replicates of 15–20 larvae. (C–E) ppk > +n = 14 neurons, ppk > Subd RNAi n = 12 neurons. *p < 0.05, **p < 0.01.

Figure 6.

Figure 6—figure supplement 1. Subdued mutants have reduced larvae growth during the second instar.

Figure 6—figure supplement 1.

(A) Subdued expression in knockout mutant larvae measured by qRT-PCR. (B) Size of second instar larvae mutant for subdued measured by body area. (C) Subdued expression in Subdued-RNAi expressing larvae measured by qRT-PCR. (D) Size of larvae expressing subdued-RNAi in Class IV dendritic arborization (C4da) neurons. (E) Ecdysone regulates subdued expression to regulate the sC4da neuron sensory switch. (A) Student’s t-test. (BD) One-way analysis of variance (ANOVA). *p < 0.05, **p < 0.01. (A, C) n = 3 whole larvae isolation replicates.

Having found that UV light-induced Ca2+ activity in C4da neurons decreased during larval development, we sought to determine whether Subdued is involved in the reduction of UV-A Ca2+ activity from second to third instar. We observed that expression of subdued-RNAi decreased both the number of actively responding neurons and the peak Ca2+ activity in C4da neurons during light stimulation of second instar larvae (Figure 6C–E).

Together these data demonstrate that reducing expression of subdued in second instar larvae is sufficient not only to increase thermal nociception but also to decrease the UV-A light response, thus enabling second instar C4da neurons to respond to thermal as well as light stimuli in the same manner as third instar C4da neurons. These effects of reduced subdued expression match the developmental progression of the sensory switch. We conclude that EcR-A regulation of the C4da sensory switch is mediated in part by reduction of subdued expression levels: subdued expression, owing to the EcR-A unliganded activity in second instar C4da neurons, is associated with a lack of thermal nociception, whereas suppression of subdued expression via EcR-A overexpression or subdued-RNAi in second instar C4da neurons causes precocious thermal nociception while reducing responsiveness to UV light (Figure 6—figure supplement 1E).

Discussion

The impact of peripheral nervous system (PNS) development on behavior is underscored by the recent finding that defects in the PNS are sufficient to produce symptoms in mouse models of autism spectrum disorders (Orefice et al., 2016). The role of hormonal regulation of sensory systems during periods of postnatal development has also been highlighted by studies of growth hormone regulation of thermal sensitivity (Liu et al., 2017; Ford et al., 2019). Considering that sensory neurons progress through transcriptionally distinct identities during development (Sharma et al., 2020), characterizing how the specific changes in receptors or channels that produce these developmentally specific outcomes is of prominent importance for understanding the development of neurological disease.

In this study, we address how a nociceptive behavior is temporally acquired during development. We demonstrate how steroid hormone regulation of nociceptor activity in a class of sensory neurons in the PNS adjusts the developing larval response to nociceptive stimuli. Heat-induced nociceptive behavior in D. melanogaster larvae arises during the final (third) instar of development (Sulkowski et al., 2011). Here, we show that this behavioral transition reflects a nociceptive sensory switch through regulation of Subdued by the steroid hormone ecdysone and the EcR isoform EcR-A.

While previous studies have focused on EcR localization in the third instar and pupal stages of C4da neurons, we show EcR is present in both the nucleus and cytoplasm in the second instar, followed with an increase of EcR nuclear localization 8 hr into the third instar (Figure 2). Previous work has shown that loss of EcR-A reduces thermal nociception in the third instar (McParland et al., 2015). While we did not find detectable transcriptional changes in nociceptor genes in third instar C4da neurons during EcR-A RNAi expression, we found broad suppression of nociceptor genes during EcR ligand-binding mutant (W650A) expression (Figure 5—figure supplement 2). Through behavioral and transcriptional analysis, we show that EcR-A nociception regulation is ligand dependent, by demonstrating that increased ecdysone titer and overexpression of ligand-competent EcR both accelerate nociception in the second instar (Figures 3 and 4). Ligand binding of EcR leads to widespread epigenetic changes through release of corepressors and recruitment of coactivators (Uyehara et al., 2017; Uyehara and McKay, 2019). The precocious activation of nociception in the second instar requires the unique A/B domain of the isoform EcR-A, as other isoforms lacking this domain do not increase nociception (Figure 4A). The A/B domain of EcR-A is less activating and more repressive then the EcR-B1 A/B domain (Dela Cruz et al., 2000; Mouillet et al., 2001; Hu et al., 2003). It thus appears that ligand-dependent derepression could be the action which promotes nociception. Moreover, expression of the EcR coactivator mutant (F645A) phenocopies expression of wild-type EcR in its ability to promote nociception (Figure 4B), suggesting that in contrast to coactivator recruitment, preservation of the ability to remove corepressors is required for nociception. This regulation of repression may account for the ability of the EcR ligand-binding mutant (W650A) to suppress multiple nociceptive genes including subdued in the second and third instars. The release of corepressors could allow other transcription factors to gain access to response elements which suppress subdued, or directly promote transcription of a repressor of subdued. How derepression leads to the suppression of subdued is an intriguing open question.

For third instar larval nociception, our data suggest there are both ligand-dependent and -independent EcR-A pathways. We find third instar expression of EcR-A ligand-binding mutant to have a temperature-specific phenotype. Expression of the EcR-A ligand-binding mutant inhibits nociception at 42°C, but does not reduce nociception at 46°C (Figure 3B, D). This temperature-specific effect is reminiscent of the phenotype of TrpA1 mutant larvae, which lose nociceptive behavior at probe temperatures of 44°C and lower, but are still responsive to 46°C stimuli (Gu et al., 2019). We found that expression of EcR-A ligand-binding mutant reduced subdued expression in the third instar even more than during the developmental transition (Figure 5A, C), suggesting that this loss of subdued expression may contribute to the phenotype at 42°C but not at 46°C. Consistent with this possibility, we observed that Subdued-RNAi was able to increase nociception in second instar larvae while Subdued knockout mutation did not increase nociception. Together these results suggest that the amount of subdued expression is important for promoting or inhibiting nociception. Furthermore, there are Subdued and EcR-A ligand-independent pathways which promote nociception in the third instar. We observe alteration of TrpA1, painless, ppk1 and ppk26 expression during development but not as a result of EcR overexpression (Figure 5). These separate pathways may involve mechanisms of regulating temperature specificity in C4da neurons.

Subdued has homology to the Calcium-activated Chloride Channels (CaCC) of the mammalian TMEM16 family, with channel activity similar to both TMEM16A and TMEM16F (Wong et al., 2013; Le et al., 2019). The effect of CaCC on neural activity is dependent on the intracellular Cl concentration ([Cl]i) (Berg et al., 2012). When [Cl]i is high, elevation of internal Ca2+ level activates CaCCs and causes Cl efflux, leading to excitation. In contrast, when [Cl]i is low, Ca2+ activation of CaCCs causes an influx of Cl which has an inhibitory effect on neuronal excitability. Differential expression of symporters and cotransporters in the CNS and PNS, as well as in immature and mature neurons, is important for creating the inhibitory or excitatory effects of CaCCs. In third instar larvae, Cl efflux has been observed during thermal nociceptive stimulation of C4da neurons (Onodera et al., 2017). However, Onodera et al. found that loss of Subdued did not significantly change the magnitude of the Cl efflux during heat stimulus, suggesting other channels are responsible for the Cl efflux in third instar C4da neurons. The direction of Cl movement in the second instar remains to be determined, and it is conceivable that Subdued could produce an inhibitory influx of Cl in the second instar. Characterizing Cl homeostasis during C4da neuron development and its impact on nociception will be of interest in future studies.

Subdued has previously been found to regulate thermal nociception in third instar larvae (Jang et al., 2015). Jang et al. found the strongest decrease of nociception in the third instar when expression of subdued was reduced in all dendritic arborization neurons. Additionally, when subdued was overexpressed in C4da neurons the number of third instar larvae that responded to nociceptive heat did not significantly change. Interesting, when subdued was overexpressed with a subdued-Gal4, the number of fast responding third instar larvae increased (Jang et al., 2015). This suggests there may be a substantial thermal nociception role for Subdued in dendritic arborization neurons other than C4da neurons during third instar. Our data suggest that as C4da neurons develop from second to third instars, the ability of Subdued to influence detection of stimuli may change. We find that thermal nociception is enhanced by reduction of subdued expression in second instar C4da neurons, via either EcR-A overexpression or Subdued-RNAi knockdown. Additionally, we find that expression of the EcR-A-ligand mutant reduces subdued expression even further in third instar C4da neurons, and that this is associated with a loss of nociception. The greater amount of subdued expression in the second instar may mean that Subdued could have a greater impact on Cl regulation during the second instar. This mechanism of Subdued activity would be dependent on the intracellular Cl concentration created by symporters and cotransporters. The importance of Cl regulation has recently been highlighted in the Class III da neurons, as both Subdued and depolarizing Cl currents have recently been found to promote cold-evoked nociception (Himmel et al., 2021). Therefore, identifying regulators of Cl concentration during the development of C4da neurons and their impact on nociception will be of interest in future studies.

Our work adds to the evidence that the developmental transition of thermal nociception represents a shift in sensory state rather than a change in sensory system construction: (1) optogenetics can stimulate aversive behavior in both second and third instar C4da neurons (Sulkowski et al., 2011), (2) TrpA1 misexpression can confer nociceptive behavior in both second and third instar larvae exposed to innocuous temperatures (Luo et al., 2017), (3) UV-A stimulation induces greater Ca2+ response in second instar than in third instar C4da neurons (Figure 1), and (4) reduction of subdued expression in second instar renders C4da neurons responsive to thermal nociceptive stimuli (Figure 6). Additional properties of C4da neurons during this developmental transition remain to be determined. These properties include: whether thermally induced calcium activity in C4da neurons (Terada et al., 2016; Gu et al., 2019) also changes over this period of development, whether EcR regulates the ‘burst and pause’ encoding (Terada et al., 2016), and whether EcR regulation alters synaptic connections of C4da neurons (Valdes-Aleman et al., 2021).

Recent findings have highlighted the occurrence of developmentally timed modulation of sensory state in both larvae and adult Drosophila. A sensory switch in thermotaxis behavior of Drosophila larvae is regulated by transcriptional regulation of thermoreceptors (Tyrrell et al., 2021). In adults, male courtship behavior is modulated through olfactory neuron pheromone sensitivity and hormone-mediate chromatin reprograming (Zhao et al., 2020). Our study highlights that temporally programed sensory switches may be a widely used developmental mechanism to match behavioral outcomes with life history events.

We speculate that the nociceptive shift in sensory state is an important mechanism for matching sensory system function with life history transitions. D. melanogaster larvae do not develop robust thermal nociception until the second half of the larval phase. This means that for the first half of larval life, larvae do not have thermal nociception behavior. In contrast, nociceptor activity to UV-A light decreases from the second to third instars, suggesting that nociceptive sensitivity to short wavelength light decreases during the second half of the larval phase. This transition correlates with larval behavioral changes that emerge during the third instar, such as cessation of feeding beneath the surface and wandering above ground (Wegman et al., 2010). Larvae are transitioning from predominantly feeding behavior in the second instar, often shielded from light and high temperatures, to movement away from food in search of pupation sites in the third instar. In a natural habitat, this transition can mean leaving the food source, which necessitates greater exposure to heat, increasing the risk of predation and desiccation (Ballman et al., 2017). This sensory switch may function as a developmental strategy, shaping behavioral outcomes as animals encounter changing environments, thus increasing survival.

Materials and methods

D. melanogaster stocks and culturing

Detailed stock genotypes and sources are listed in Table 1. X chromosome genotypes are simplified: male and female larvae are pooled together in test populations, and the origin of miniwhite alleles could be mixed. Control genotypes are crossed to w1118 (v60000) unless otherwise noted.

Table 1. Drosophila melanogaster stocks used in this study.

Figure Genotype Source
1A and S1-1 ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+ Han et al., 2011
1C–F ppk-Gal4vk37/+;UAS-GCaMP6(s),UAS-tdTom/+ BDSC 42749, Han et al., 2011
2A–C, S2-1B, 3A ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
3B and S3-1A–C ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
P{UAS-EcR.A.W650A}TP5/+ BDSC 9451
P{UAS-EcR.A.F645A}TP2/+ BDSC 9452
ppk-tdGFP1b,UAS-Dcr2/P{UAS-EcR.A.F645A}TP2;ppk-Gal41a/+
ppk-tdGFP1b,UAS-Dcr2/P{UAS-EcR.A.W650A}TP5;ppk-Gal41a/+
4A and S4-2A, B ppk-Gal4vk37 Han et al., 2011
P{UAS-EcR.A}3a/+ BDSC 6470
P{UAS-EcR.B1}3b/+ BDSC 6469
P{UAS-EcR.C}TP1-4/+ BDSC 6868
ppk-Gal4vk37;P{UAS-EcR.A}3a
ppk-Gal4vk37; P{UAS-EcR.B1}3b
ppk-Gal4vk37;P{UAS-EcR.C}TP1-4
4B and S4-1A–G ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
P{UAS-EcR.A}3a/+
P{UAS-EcR.A.F645A}TP2/+
P{UAS-EcR.A.W650A}TP5/+
ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/P{UAS-EcR.A}3a
ppk-tdGFP1b,UAS-Dcr2/P{UAS-EcR.A.F645A}TP2;ppk-Gal41a/+
ppk-tdGFP1b,UAS-Dcr2/P{UAS-EcR.A.W650A}TP5;ppk-Gal41a/+
5A ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
5B ppk-CD4- tdGFP1b /+
ppk-CD4- tdGFP1b /+;P{UAS-EcR.A}3a/+
ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/P{UAS-EcR.A}3a
S5-1A, B ppk-CD4- tdGFP1b /+
ppk-CD4- tdGFP1b / P{UAS-EcR.A.W650A}TP5
ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
ppk-tdGFP1b,UAS-Dcr2/P{UAS-EcR.A.W650A}TP5;ppk-Gal41a/+
S5-2A, B ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
ppk-CD4- tdGFP1b /+;P{w[ + mC] = UAS EcR.A.dsRNA}91/+ BDSC 9328
ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/ P{w[ + mC] = UAS EcR.A.dsRNA}91
6A, S6-1A, B SubduedKO11/+ Wong et al., 2013
Df(3R)Exel6184, P{w[ + mC] = XPU}Exel6184 BDSC 7663
SubduedKO11/Df(3R)Exel6184, P{w[ + mC] = XPU}Exel6184
6B, S6-1D ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/+
P{GD3674}v37472/+ VDRC 37472
ppk-tdGFP1b,UAS-Dcr2/+;ppk-Gal41a/ P{GD3674}v37472
S6-1C P{GD3674}v37472/+ VDRC 37472
Actin-Gal4/+; SubduedKO2/+ Wong et al., 2013
Actin-Gal4/+; SubduedKO2/P{GD3674}v37472
6-1C–E ppk-Gal4vk37/+;UAS-GCaMP6(s),UAS-tdTom/+ BDSC 42749, Han et al., 2011
ppk-Gal4vk37/+;UAS-GCaMP6(s),UAS-tdTom/ P{GD3674}v37472

All experimental crosses were raised at 25°C and 60% humidity, with a 12 hr light–dark cycle, and fed cornmeal-molasses media. Larval development was synchronized by staged egg collection on grape agar, followed by synchronized collection of newly hatched L1 larvae. Second instar larvae were assayed between 26 and 32 hr after L1 larval hatching and third instar larvae were collected as prepupariation wandering larvae. All assays were performed during the dark–light cycle.

Nociceptive behavior assay

Nociceptive stimulation was performed as previously described (Babcock et al., 2009). Larvae were briefly rinse in dH20 and allowed to acclimatize for 1 min on a vinyl sheet. The larvae were kept moist and a temperature-controlled heat probe (ProDev Engineering, TX) was used to apply heat between the larval body segments A3–A6 of forward crawling larvae. Nociceptive behavior was defined as at least one complete 360° roll. The time to complete one roll (latency) was measured up to a 20 s cutoff. Each larva was presented with a single stimulus to avoid habituation. Latency was calculated for each biological replicate as the time at which greater than or equal to 50% of the responding population (nonresponding larvae we excluded) had responded to the stimulus.

20E feeding

For food supplemented with 20E (Sigma, H5142), 20E was dissolved in 100% ethanol and mixed with room temperature cornmeal-molasses media. An equivalent volume of ethanol alone was mixed with media as a control. Larvae were transferred to supplemented media 48 hr AEL and assayed for nociceptive behavior 8 hr later.

Ultraviolet light response calcium imaging

C4da response to UV-A stimulation was measured as previously described (Yadav et al., 2019). A Leica SP5 confocal microscope with resonance scanner was used with the ×20 oil immersion objective and ×16 optical magnification. Imaging was done with 512 × 512 resolution and a slice thickness of 5 μm (20 Z slices) at 1.201 fps. A 405 laser line (50 mW) was used at 100% laser power. Neurons were imaged for 60 s before a 10-s UV exposure. Z-stacks were acquired in the GFP, RFP, an UV channels. The ratiometric signal was quantified as described for thermal Ca2+ imaging.

Cell purification and qRT-PCR

C4da neurons were isolated by dissecting 30–40 larvae in 1× phosphate-buffered saline (PBS) on ice. To increase C4da neuron concentration in the final cell suspension, the larval body wall was inverted and the CNS, imaginal discs, gut, and fatbody were removed. After dissection cell suspensions were prepared by vortexed with 1.5 μl 1× Liberase TM (Roche, LIBTM-RO) in 500 μl cold PBS. To dissociate cells samples were incubated for three periods at 25°C at 1000 rpm, triturating 10 times with a glass pipette between each incubation. Suspensions were strained through a 40-μm cell strainer (Fisher Scientific) and brought to 1.5 ml with Schneider’s media. 1 μl ethidium homodimer-1 (Thermo Fisher, L3224) was added before sorting to mark dead cells. C4da neurons were isolated by FACS with an Aria II (Becton Dickinson). GFP+nonautofluorescent RFP− events were sorted into 20 μl lysis buffer (Thermo Fisher, KIT0204) and immediately frozen on dry ice. 100–1000 cells were isolated per dissection replicate. RNA was isolated with PicoPure RNA isolation kit (Thermo Fisher, KIT0204), cDNA was synthesized with Applied Biosystems High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher, 4368813), and qRT-PCR was performed using SYBR green (Thermo Fisher, A25742) with a QuantStudio7 (Thermo Fisher). Relative expression was calculated using the detla-delta Ct method with the housekeeping gene eEF1α2 normalized to the mean Ct of the genotypic and isolation controls.

Subdued expression in knockout mutant and RNAi expressing larvae was measured using RNeasy Mini Kit (Qiagen 74104) isolation from three to four larvae per genotype. cDNA syntheses and qRT-PCR were performed as described FACS isolated cells.

Primers

Target Forward Reverse
TrpA1 isoformCD GCCGGAACAGCAAGTATTGGA CGATTTCAATCCGCTTGGGAC
painless CAGTGAGCGACACCCAAGTTA GAGAACCTGCTCGTACCGAC
subdued GGAGGTCGAGTCCAGTCAGA CTGGAATCTCCTTCATGGGCA
ppk1 CCCGAAAAACGCCAATGTCT ACAACCGCATTTGGAAACCG
ppk26 GAGCGGAAGGTATTATTCCCGA GGGAGATGTATCCGCACTGG
piezo AAGCCACGGGTTTCTTTGC GTTGGGGTGTACGTGCCTTT
eEF1α2 CGTCTACAAGATCGGAGGCA GACCATGCCTGGCTTGAGGA

Immunohistochemistry

Larval ages were staged by L1/L2 molt for second instar or L2/L3 molt for third instar. Larvae were filleted in PBS on ice and fixed for 10 min at room temperature in 4% paraformaldehyde. Samples were blocked with 5% goat serum for 2 hr at room temperature and then incubated with either EcR-common DDA2.7, EcR-A 15G1a, or EcR-B1 AD4.4 (Developmental Studies Hybridoma Bank) in blocking solution (1:200) overnight at 4°C. After rinse, samples were incubated with secondary antibody Alexa Fluor 555 (Invitrogen A-2142A) in blocking solution (1:500) for 2 hr at room temperature before 10 min staining with DAPI (1:10,000) and mounting in Vectashield (Vector Laboratories). Mean fluorescence was measured with Fiji (https://imagej.net/fiji) in traces of the nuclei and cytoplasm (traces of cytoplasm excluded C4da neuron nuclei and an neighboring nuclei) and adjusted for area before calculating the ratio of fluorescence in the nuclease and cytoplasm of C4da neurons.

Larval size measurement

Larvae were collected in 75 μl PBS and placed in an 80°C thermo-block for 10 min. The turgid larvae were then placed on a microscope slide, imaged, and larval body area was measured with Fiji (https://imagej.net/fiji).

Statistical analysis

Statistical tests were done with GraphPad Prism 8.3.0 (GraphPad Software) or R (version 3.5.2, R Core Team 2018).

Acknowledgements

We thank Tun Li and Susan Younger for experimental consultation, Caitlin O’Brien, Han-Hsuan Liu, Ke Li, Maja Petkovic, Beverly Piggott, and Rebecca Jaszczak for editorial advice. Stocks obtained from the Bloomington Drosophila Stock Center (BDSC, NIH P40OD018537) and Vienna Drosophila Resource Center (VDRC, https://www.vdrc.at) were used in this study. DDA27 was obtained from the from the Developmental Studies Hybridoma Bank (DSHB). Research reported in this publication was supported by the National Institutes of Health: National Institute of General Medical Sciences F32GM130019 (JSJ) and National Institute of Neurological Disorders and Stroke R35NS097227 (YNJ). YNJ and LYJ are investigators at the Howard Hughes Medical Institute.

Funding Statement

The funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication.

Contributor Information

Yuh Nung Jan, Email: YuhNung.Jan@ucsf.edu.

Mani Ramaswami, Trinity College Dublin, Ireland.

K VijayRaghavan, National Centre for Biological Sciences, Tata Institute of Fundamental Research, India.

Funding Information

This paper was supported by the following grants:

  • National Institute of General Medical Sciences F32GM130019 to Jacob S Jaszczak.

  • National Institute of Neurological Disorders and Stroke R35NS097227 to Yuh Nung Jan.

  • Howard Hughes Medical Institute to Lily Yeh Jan, Yuh Nung Jan.

Additional information

Competing interests

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Validation, Visualization, Writing - original draft, Writing - review and editing.

Data curation, Investigation, Methodology, Validation, Writing - review and editing.

Funding acquisition, Writing - review and editing.

Funding acquisition, Methodology, Supervision, Writing - review and editing.

Additional files

Transparent reporting form

Data availability

All data generated or analyzed during this study are included in the manuscript and supporting files.

References

  1. Ainsley JA, Kim MJ, Wegman LJ, Pettus JM, Johnson WA. Sensory mechanisms controlling the timing of larval developmental and behavioral transitions require the Drosophila DEG/ENaC subunit, Pickpocket1. Developmental Biology. 2008;322:46–55. doi: 10.1016/j.ydbio.2008.07.003. [DOI] [PubMed] [Google Scholar]
  2. Almeida-Carvalho MJ, Berh D, Braun A, Chen Y, Eichler K, Eschbach C, Fritsch PMJ, Gerber B, Hoyer N, Jiang X, Kleber J, Klämbt C, König C, Louis M, Michels B, Miroschnikow A, Mirth C, Miura D, Niewalda T, Otto N, Paisios E, Pankratz MJ, Petersen M, Ramsperger N, Randel N, Risse B, Saumweber T, Schlegel P, Schleyer M, Soba P, Sprecher SG, Tanimura T, Thum AS, Toshima N, Truman JW, Yarali A, Zlatic M. The Ol1mpiad: Concordance of behavioural faculties of stage 1 and stage 3 Drosophila larvae. Journal of Experimental Biology. 2017;220:2452–2475. doi: 10.1242/jeb.156646. [DOI] [PubMed] [Google Scholar]
  3. Arenas OM, Zaharieva EE, Para A, Vásquez-Doorman C, Petersen CP, Gallio M. Activation of planarian TRPA1 by reactive oxygen species reveals a conserved mechanism for animal nociception. Nature Neuroscience. 2017;20:1686–1693. doi: 10.1038/s41593-017-0005-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babcock DT, Landry C, Galko MJ. Cytokine Signaling Mediates UV-Induced Nociceptive Sensitization in Drosophila Larvae. Current Biology. 2009;19:799–806. doi: 10.1016/j.cub.2009.03.062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baliki MN, Apkarian AV. Nociception, Pain, Negative Moods, and Behavior Selection. Neuron. 2015;87:474–491. doi: 10.1016/j.neuron.2015.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Ballman ES, Collins JA, Drummond FA. Pupation Behavior and Predation on Drosophila suzukii (Diptera: Drosophilidae) Pupae in Maine Wild Blueberry Fields. Journal of Economic Entomology. 2017;110:2308–2317. doi: 10.1093/jee/tox233. [DOI] [PubMed] [Google Scholar]
  7. Berg J, Yang HH, Jan LY. Ca2+-activated Cl- channels at a glance. Journal of Cell Science. 2012;125:1367–1371. doi: 10.1242/jcs.093260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cherbas L, Hu X, Zhimulev I, Belyaeva E, Cherbas P. EcR isoforms in Drosophila: Testing tissue-specific requirements by targeted blockade and rescue. Development (Cambridge, England) 2003;130:271–284. doi: 10.1242/dev.00205. [DOI] [PubMed] [Google Scholar]
  9. Colombani J, Bianchini L, Layalle S, Pondeville E, Dauphin-Villemant C, Antoniewski C, Carré C, Noselli S, Léopold P. Antagonistic actions of ecdysone and insulins determine final size in Drosophila. Science (New York, N.Y.) 2005;310:667–670. doi: 10.1126/science.1119432. [DOI] [PubMed] [Google Scholar]
  10. Dela Cruz FE, Kirsch DR, Heinrich JN. Transcriptional activity of Drosophila melanogaster ecdysone receptor isoforms and ultraspiracle in Saccharomyces cerevisiae. Journal of Molecular Endocrinology. 2000;24:183–191. doi: 10.1677/jme.0.0240183. [DOI] [PubMed] [Google Scholar]
  11. Ford ZK, Dourson AJ, Liu X, Lu P, Green KJ, Hudgins RC, Jankowski MP. Systemic growth hormone deficiency causes mechanical and thermal hypersensitivity during early postnatal development. IBRO Reports. 2019;6:111–121. doi: 10.1016/j.ibror.2019.02.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gorczyca DA, Younger S, Meltzer S, Kim SE, Cheng L, Song W, Lee HY, Jan LY, Jan YN. Identification of Ppk26, a DEG/ENaC Channel Functioning with Ppk1 in a Mutually Dependent Manner to Guide Locomotion Behavior in Drosophila. Cell Reports. 2014;9:1446–1458. doi: 10.1016/j.celrep.2014.10.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gu P, Gong J, Shang Y, Wang F, Ruppell KT, Ma Z, Sheehan AE, Freeman MR, Xiang Y. Polymodal Nociception in Drosophila Requires Alternative Splicing of TrpA1. Current Biology. 2019;29:3961–3973. doi: 10.1016/j.cub.2019.09.070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Han C, Jan LY, Jan YN. Enhancer-driven membrane markers for analysis of nonautonomous mechanisms reveal neuron-glia interactions in Drosophila. PNAS. 2011;108:9673–9678. doi: 10.1073/pnas.1106386108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hill RJ, Billas IML, Bonneton F, Graham LD, Lawrence MC. Ecdysone Receptors: From the Ashburner Model to Structural Biology. Annual Review of Entomology. 2013;58:251–271. doi: 10.1146/annurev-ento-120811-153610. [DOI] [PubMed] [Google Scholar]
  16. Himmel NJ, Patel AA, Cox DN, Himmel NJ, Patel AA, Cox DN. Invertebrate Nociception. Oxford Research Encyclopedia of Neuroscience. 2017;1:29. doi: 10.1093/acrefore/9780190264086.013.166. [DOI] [Google Scholar]
  17. Himmel NJ, Sakurai A, Letcher JM, Patel AA, Bhattacharjee S, Benson MN, Gray TR, Cymbalyuk GS, Cox DN. Chloride-dependent mechanisms of multimodal sensory discrimination and neuropathic sensitization in Drosophila. Neuroscience. 2021;10:472093. doi: 10.1101/2021.12.10.472093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hiura A, López Villalobos E, Ishizuka H. Age-dependent attenuation of the decrease of c fibers by capsaicin and its effects on responses to nociceptive stimuli. Somatosensory & Motor Research. 1992;9:37–43. doi: 10.3109/08990229209144761. [DOI] [PubMed] [Google Scholar]
  19. Honjo K, Mauthner SE, Wang Y, Skene JHP, Tracey WD. Nociceptor-Enriched Genes Required for Normal Thermal Nociception. Cell Reports. 2016;16:295–303. doi: 10.1016/j.celrep.2016.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hu X, Cherbas L, Cherbas P. Transcription Activation by the Ecdysone Receptor (EcR/USP): Identification of Activation Functions. Molecular Endocrinology (Baltimore, Md.) 2003;17:716–731. doi: 10.1210/me.2002-0287. [DOI] [PubMed] [Google Scholar]
  21. Hwang RY, Zhong L, Xu Y, Johnson T, Zhang F, Deisseroth K, Tracey WD. Nociceptive Neurons Protect Drosophila Larvae from Parasitoid Wasps. Current Biology. 2007;17:2105–2116. doi: 10.1016/j.cub.2007.11.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jang W, Kim JY, Cui S, Jo J, Lee BCC, Lee Y, Kwon KSS, Park CSS, Kim C. The anoctamin family channel subdued mediates thermal nociception in Drosophila. The Journal of Biological Chemistry. 2015;290:2521–2528. doi: 10.1074/jbc.M114.592758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jankowski MP, Ross JL, Weber JD, Lee FB, Shank AT, Hudgins RC. Age-dependent sensitization of cutaneous nociceptors during developmental inflammation. Molecular Pain. 2014;10:1–18. doi: 10.1186/1744-8069-10-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kannangara JR, Mirth CK, Warr CG. Regulation of ecdysone production in Drosophila by neuropeptides and peptide hormones. Open Biology. 2021;11:200373. doi: 10.1098/rsob.200373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kim SE, Coste B, Chadha A, Cook B, Patapoutian A. The role of Drosophila Piezo in mechanical nociception. Nature. 2012;483:209–212. doi: 10.1038/nature10801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kirilly D, Gu Y, Huang Y, Wu Z, Bashirullah A, Low BC, Kolodkin AL, Wang H, Yu F. A genetic pathway composed of Sox14 and Mical governs severing of dendrites during pruning. Nature Neuroscience. 2009;12:1497–1505. doi: 10.1038/nn.2415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kuo CT, Jan LY, Jan YN. Dendrite-specific remodeling of Drosophila sensory neurons requires matrix metalloproteases, ubiquitin-proteasome, and ecdysone signaling. PNAS. 2005;102:15230–15235. doi: 10.1073/pnas.0507393102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Le T, Le SC, Yang H. Drosophila Subdued is a moonlighting transmembrane protein 16 (TMEM16) that transports ions and phospholipids. The Journal of Biological Chemistry. 2019;294:4529–4537. doi: 10.1074/jbc.AC118.006530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Liu L, Yermolaieva O, Johnson WA, Abboud FM, Welsh MJ. Identification and function of thermosensory neurons in Drosophila larvae. Nature Neuroscience. 2003;6:267–273. doi: 10.1038/nn1009. [DOI] [PubMed] [Google Scholar]
  30. Liu X, Green KJ, Ford ZK, Queme LF, Lu P, Ross JL, Lee FB, Shank AT, Hudgins RC, Jankowski MP. Growth hormone regulates the sensitization of developing peripheral nociceptors during cutaneous inflammation. Pain. 2017;158:333–346. doi: 10.1097/j.pain.0000000000000770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Luo J, Shen WL, Montell C. TRPA1 mediates sensation of the rate of temperature change in Drosophila larvae. Nature Neuroscience. 2017;20:34–41. doi: 10.1038/nn.4416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mauthner SE, Hwang RY, Lewis AH, Xiao Q, Tsubouchi A, Wang Y, Honjo K, Skene J, Grandl J, Tracey WD. Balboa binds to pickpocket in vivo and is required for mechanical nociception in Drosophila larvae. Current Biology. 2014;24:2920–2925. doi: 10.1016/j.cub.2014.10.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. McParland AL, Follansbee TL, Vesenka GD, Panaitiu AE, Ganter GK. Steroid receptor isoform expression in Drosophila nociceptor neurons is required for normal dendritic arbor and sensitivity. PLOS ONE. 2015;10:140785. doi: 10.1371/journal.pone.0140785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mouillet JF, Henrich VC, Lezzi M, Vögtli M. Differential control of gene activity by isoforms A, B1 and B2 of the Drosophila ecdysone receptor. European Journal of Biochemistry. 2001;268:1811–1819. [PubMed] [Google Scholar]
  35. Nieva C, Spindler-Barth M, Azoitei A, Spindler KD. Influence of hormone on intracellular localization of the Drosophila melanogaster ecdysteroid receptor (EcR) Cellular Signalling. 2007;19:2582–2587. doi: 10.1016/j.cellsig.2007.08.007. [DOI] [PubMed] [Google Scholar]
  36. Nieva C, Spindler-Barth M, Spindler KD. Impact of heterodimerization on intracellular localization of the Ecdysteroid Receptor (EcR) Archives of Insect Biochemistry and Physiology. 2008;68:40–48. doi: 10.1002/arch.20234. [DOI] [PubMed] [Google Scholar]
  37. Onodera K, Baba S, Murakami A, Uemura T, Usui T. Small conductance Ca2+-activated K+ channels induce the firing pause periods during the activation of Drosophila nociceptive neurons. eLife. 2017;6:e29754. doi: 10.7554/eLife.29754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Orefice LLL, Zimmerman ALL, Chirila AMM, Sleboda SJJ, Head JPP, Ginty DDD. Peripheral Mechanosensory Neuron Dysfunction Underlies Tactile and Behavioral Deficits in Mouse Models of ASDs. Cell. 2016;166:299–313. doi: 10.1016/j.cell.2016.05.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ou Y, Chwalla B, Landgraf M, van Meyel DJ. Identification of genes influencing dendrite morphogenesis in developing peripheral sensory and central motor neurons. Neural Development. 2008;3:1–27. doi: 10.1186/1749-8104-3-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Parrish JZ, Kim CC, Tang L, Bergquist S, Wang T, Derisi JL, Jan LY, Jan YN, Davis GW. Kruppel Mediates the Selective Rebalancing of Ion Channel Expression. Neuron. 2014;82:537–544. doi: 10.1016/j.neuron.2014.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Rewitz KF, Yamanaka N, O’Connor MB. Developmental Checkpoints and Feedback Circuits Time Insect Maturation. Current Topics in Developmental Biology. 2013;103:1–33. doi: 10.1016/B978-0-12-385979-2.00001-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Schubiger M, Tomita S, Sung C, Robinow S, Truman JW. Isoform specific control of gene activity in vivo by the Drosophila ecdysone receptor. Mechanisms of Development. 2003;120:909–918. doi: 10.1016/s0925-4773(03)00134-5. [DOI] [PubMed] [Google Scholar]
  43. Sharma N, Flaherty K, Lezgiyeva K, Wagner DE, Klein AM, Ginty DD. The emergence of transcriptional identity in somatosensory neurons. Nature. 2020;577:392–398. doi: 10.1038/s41586-019-1900-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sokabe T, Tsujiuchi S, Kadowaki T, Tominaga M. Drosophila painless is a Ca2+-requiring channel activated by noxious heat. The Journal of Neuroscience. 2008;28:9929–9938. doi: 10.1523/JNEUROSCI.2757-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sulkowski MJ, Kurosawa MS, Cox DN. Growing pains: Development of the larval nocifensive response in Drosophila. The Biological Bulletin. 2011;221:300–306. doi: 10.1086/BBLv221n3p300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sullivan AA, Thummel CS. Temporal Profiles of Nuclear Receptor Gene Expression Reveal Coordinate Transcriptional Responses during Drosophila Development. Molecular Endocrinology (Baltimore, Md.) 2003;17:2125–2137. doi: 10.1210/me.2002-0430. [DOI] [PubMed] [Google Scholar]
  47. Terada SI, Matsubara D, Onodera K, Matsuzaki M, Uemura T, Usui T. Neuronal processing of noxious thermal stimuli mediated by dendritic ca2+ influx in Drosophila somatosensory neurons. eLife. 2016;5:e12959. doi: 10.7554/eLife.12959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tracey WD, Wilson RI, Laurent G, Benzer S. painless, a Drosophila gene essential for nociception. Cell. 2003;113:261–273. doi: 10.1016/s0092-8674(03)00272-1. [DOI] [PubMed] [Google Scholar]
  49. Tyrrell JJ, Wilbourne JT, Omelchenko AA, Yoon J, Ni L. Ionotropic Receptor-dependent cool cells control the transition of temperature preference in Drosophila larvae ed. PLOS Genetics. 2021;17:e1009499. doi: 10.1371/journal.pgen.1009499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Uyehara CM, Nystrom SL, Niederhuber MJ, Leatham-Jensen M, Ma Y, Buttitta LA, McKay DJ. Hormone-dependent control of developmental timing through regulation of chromatin accessibility. Genes & Development. 2017;31:862–875. doi: 10.1101/gad.298182.117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Uyehara CM, McKay DJ. Direct and widespread role for the nuclear receptor EcR in mediating the response to ecdysone in Drosophila. PNAS. 2019;116:9893–9902. doi: 10.1073/pnas.1900343116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Valdes-Aleman J, Fetter RD, Sales EC, Heckman EL, Venkatasubramanian L, Doe CQ, Landgraf M, Cardona A, Zlatic M. Comparative Connectomics Reveals How Partner Identity, Location, and Activity Specify Synaptic Connectivity in Drosophila. Neuron. 2021;109:105–122. doi: 10.1016/j.neuron.2020.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Warren JT, Yerushalmi Y, Shimell MJ, O’Connor MB, Restifo LL, Gilbert LI. Discrete pulses of molting hormone, 20-hydroxyecdysone, during late larval development of Drosophila melanogaster: Correlations with changes in gene activity. Developmental Dynamics. 2006;235:315–326. doi: 10.1002/dvdy.20626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wegman LJ, Ainsley JA, Johnson WA. Developmental timing of a sensory-mediated larval surfacing behavior correlates with cessation of feeding and determination of final adult size. Developmental Biology. 2010;345:170–179. doi: 10.1016/j.ydbio.2010.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wong XM, Younger S, Peters CJ, Jan YN, Jan LY. Subdued, a TMEM16 family Ca2+ -activated Cl− channel in Drosophila melanogaster with an unexpected role in host defense. eLife. 2013;2013:e00862. doi: 10.7554/eLife.00862. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Xiang Y, Yuan Q, Vogt N, Looger LL, Jan LY, Jan YN. Light-avoidance-mediating photoreceptors tile the Drosophila larval body wall. Nature. 2010;468:921–926. doi: 10.1038/nature09576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Yadav S, Younger SH, Zhang L, Thompson-Peer KL, Li T, Jan LY, Jan YN. Glial ensheathment of the somatodendritic compartment regulates sensory neuron structure and activity. PNAS. 2019;116:5126–5134. doi: 10.1073/pnas.1814456116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yamanaka N, Romero NM, Martin FA, Rewitz KF, Sun M, O’Connor MB, Léopold P. Neuroendocrine Control of Drosophila Larval Light Preference. Science (New York, N.Y.) 2013;341:1113–1116. doi: 10.1126/science.1241210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Zhao S, Deanhardt B, Barlow GT, Schleske PG, Rossi AM, Volkan PC, Volkan PC. Chromatin-based reprogramming of a courtship regulator by concurrent pheromone perception and hormone signaling. Science Advances. 2020;6:eaba6913. doi: 10.1126/sciadv.aba6913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zhong L, Hwang RY, Tracey WD. Pickpocket Is a DEG/ENaC Protein Required for Mechanical Nociception in Drosophila Larvae. Current Biology. 2010;20:429–434. doi: 10.1016/j.cub.2009.12.057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zhong L, Bellemer A, Yan H, Ken H, Jessica R, Hwang RY, Pitt GS, Tracey WD. Thermosensory and Nonthermosensory Isoforms of Drosophila melanogaster TRPA1 Reveal Heat-Sensor Domains of a ThermoTRP Channel. Cell Reports. 2012;1:43–55. doi: 10.1016/j.celrep.2011.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]

Editor's evaluation

Mani Ramaswami 1

The authors describe in vivo analyses of an intriguing steroid-mediated development shift in the sensitivity of Drosophila larvae to noxious stimulation as they move from the L2 to the L3 instar stage. Experiments and observations presented show that the steroid hormone ecdysone regulates nociceptor activity in the peripheral nervous system by suppressing expression of a gene named subdued, which encodes a membrane protein of the TMEM16 channel family.

Decision letter

Editor: Mani Ramaswami1

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "Steroid hormone signaling activates thermal nociception during Drosophila peripheral nervous system development" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor. The reviewers have opted to remain anonymous.

Our decision has been reached after careful consultation between the reviewers. Based on these discussions and the individual reviews below, we genuinely regret to inform you that this work will not be considered further for publication in eLife in current form. However, if you choose to perform all of the large number of experiments requested by the reviewers, and if you so wish, then eLife should be happy to consider an extensively revised manuscript for review, nominally as a new submission.

The reviewers felt the problem being addressed was interesting and expressed particular interest in the role of subdued as a functional target of ECR, but had several concerns regarding both the clarity of underlying mechanisms as proposed in this manuscript and the overall level of advance over previous work (eg. McPharland et al.).

Reviewer #1:

This study by the Jan group examines the onset/development of functional thermal (and UV) nociception in Drosophila larvae. Previously, other groups had established that there are differences in nociceptive responses between second and third instar larvae (Cox) and that ecdysone receptor function is required for functional nociceptor development (Ganter). This study does a more thorough job than that previous work in documenting the stage differences and identifies a molecular player (subdued) that appears to be required for stage-specific changes in sensitivity. Overall, the data are sound and convincing and the paper is very well written. A few suggestions for improvement/clarification are offered.

1. Data organization. Some of the best data showing the stage-specific difference is in supplemental figure 1C- latency. This should really be in the main figure 1- probably as panel B. Figure 1C could benefit and would be more complete if second instar data were shown for 25 and 36 degrees.

2. The baseline calcium levels in second versus third instar is quite different- much higher in second instar looking at figure 1B. This data is essentially normalized/compared to same stage's baseline but this does raise the question: Does EcR function or subdued function shift the magnitude of response or shift the second instar (or third instar) baseline? It's a little hard to tell because Figure 3 (EcR) doesn't have stills of representative cells and Figure 6 (subdued, temperature) does not either. All we see here is the ratio data. Representative still data should be presented and whether the gene affects baseline calcium versus magnitude of change upon stimulus should be discussed.

3. Figure 2 organization: Personal preference but the 20E data currently in supplement 2B is probably more important to the main point than the third instar data (Main figure 2D).

4. Some of the discussion of the differences between second and third instar seems a bit oversimplified, especially as it refers to the EcR expression data in Figure 2 supplement. It's not really accurate to say that second instar has low EcR and third has high. Reality (from the data shown) is that nuclear EcR is high in early second and later third and low in late second and early third. There isn't really data to address intra-stage differences in responsiveness so this should be described a bit more clearly- without the current oversimplification. Adding modifiers (late second instar, early third instar) would help.

5. Analysis of subdued: Most helpful would be confirmation that this RNAi line is actually reducing subdued transcript levels, as one would predict. This would enhance confidence that there are no off-target effects. Also helpful would be the converse experiment to RNAi loss of function. Authors perform Subdued RNAi and see changes in second instar thermal responsiveness. Does overexpression of Subdued shift responsiveness in third instar?

Reviewer #2:

The manuscript deals with the basis of an intriguing shift in nociceptive behavior shown by Drosophila larvae as they move from the L2 to the L3 instar. Prior publications have shown that the shift is controlled by ecdysone and implicated the EcR-A knock-down of EcR-A levels rendered the neurons hyposensitive to noxious thermal and mechanical stimulation and also altered features of their anatomy. Reduction in EcR-B1 did not effect their responses but has anatomical ramifications. This paper extends these studies into Ca2+ data from specific da neurons and also reveals a gene, subdued, that may play a role in the change in responsiveness.

There are a number of substantial concerns:

1. The paper focuses on the differences between the L2 and L3 instars. According to the Methods, their L2 stage is in the middle of the L2 period, but the L3 stage is a wandering L3 larva which is at least 50 hours older. There are three critical ecdysone-related events that occur in this 50 hour span: the critical weight at about 8-10 hr after L3 ecdysis, the mid-L3 transition, and the ecdysone peak that initiates wandering. If the transition occurs during the transition from L2 to L3, then the "critical weight" transition would be a good candidate and L3 larvae about 12 to 18 hr after ecdysis would be more appropriate.

2. Another troubling methodological issue is that their behavioral tests were run at either 42 or 46 C while the Ca2+ measurements are done at 44 C. This becomes a concern in a situation like in Figure 2 where W650A expression is behaviorally effective at 42 C but has no effect at 46 C. We are then provided with a Ca2+ response at 44 C. Since this temp is half-way between the two used for the behavior tests, we have no way of assessing whether or not the Ca2+ response tracks the behavioral response.

3. A serious interpretational issue has to do with ignoring the complexity imposed by the EcR isoforms. Through most of the paper and figures, the authors speak of "EcR" when they really should be saying "EcR-A". This is not a trivial issue because the EcR-B isoforms have a strong hormone-independent transactivation domain at their N-terminus whereas EcR-A does not. At times when EcR-B1 is a present, the expression of wildtype EcR-A dampens the cellular response by competing with the endogenous EcR-B1 [Schubiger et al., 2003, Mech. Dev. 120:909]. Previous experiments (McPharland et al.) used RNAi to selectively knockdown different isoforms and these suggested a function for EcR-A. Ectopic expression of EcR-A alone, without EcR-B1 for comparison, can lead to wrong interpretations, especially since both isoforms are found in the da neurons (see McPharland).

4. Another problem is with the interpretation of the EcR dominant negatives (EcR-DN). The function of nuclear receptors, like EcR, is too well understood to gloss over the impact of these point mutations by saying that they either prevent hormone binding [for W650A] or "co-factor" binding [for F645A]. The W650A mutation is in the binding pocket and prevents 20E binding. This prevents the loss of co-repressors and the assembly of a co-activator complex. Therefore, this forms serves as a constitutive repressor that is not responsive to 20E. The F645A substitution, by contrast is in the helix that changes conformation after 20E binding and is one of the residues involved in co-activator assembly. EcR-F645A allows 20E binding and the conformation changes that lose the co-repressors but it cannot support the co-activation. Therefore, it ca show hormone-dependent derepression but not activation.

5. Paragraph 299: This paragraph argues that expression of EcR[-A] in the L2 suppresses subdued expression and they conclude that expression of EcR-A in the L3 will enhance subdued expression because the EcR-A w650A suppresses subdued expression in the latter. They really need to show the effects of expression of the wild-type EcR-A receptor in the L3 as well as W650A. This information is important because as noted above, ectopic expression of EcR-A could act to suppress expression although likely not to the level seen for the W650A dominant negative.

6. The interpretation of the W650A data in the L3 is confusing because its expression in the third suppresses both TrpA1 and subdued [Figure 4C] but has minimal effect on Ca2+ responses to heat [Figure 3] and has no effect on behavioral responses to 46 C [Figure 2D]. I was hoping that the Discussion would pull things together into a consistent story, but the Discussion tended to invoke a number of ad hoc hypotheses to account for pieces of data which do not seem to fit.

Reviewer #3:

The manuscript by Jaszczak et al. examines how steroid hormone signaling in Drosophila activates in sensory switch in C4da neurons that promotes a thermal nociceptive escape response behavior. The authors show that a greater percent of third instar larvae are able to generate the thermal nociceptive escape response than second instar larvae. They also show calcium responses in the soma of C4da neurons are larger in third instar larvae in response to high temperatures in comparison to second instar larvae. In contrast, they see the opposite with UV light. Both the behavioral response and the calcium response can be increased in second instar larvae by increasing ecdysone signaling. The authors also find that subdued expression is suppressed by increasing ecdysone signaling and link subdued to this developmental transition. Overall, the findings in the manuscript are interesting, but I think the authors need to address several points to strengthen their conclusions.

1. Some explanation as to why the soma is being imaged instead of dendrites or axons is needed. I assume the size of dendrites and axons make it too difficult to image in intact larvae.

2. The authors mention that other groups have shown that EcR regulates dendritic growth. Could the role of EcR in this sensory switch be due to changes in dendritic growth instead of changes in calcium levels. This possibility should at least be discussed.

3. The authors use a dominant-negative approach (EcR-W650A) to reduce ecdysone signaling in third instar larvae. Dominant-negative approaches are difficult to interpret. The authors should use an EcR RNAi to knockdown ECR in C4da neurons. This should be done for both the thermal nociceptive escape responses and for the calcium imaging experiments.

4. Why does the approach dominant-negative (EcR-W650A) reduce subdued expression in third instar larvae? I would expect that reduced ecdysone signaling would enhance subdued expression.

5. TrpA1 levels in second instar larvae are not affected by increased ecdysone signaling (overexpression of EcR) in second instar larvae, but they are reduced by expression of a dominant-negative (EcR-W650A). Again, this experiment should be repeated with an EcR RNAi for third instar larvae.

6. If TrpA1 levels are regulated by ecdysone, does reducing TrpA1 in C4da neurons in third instar larvae reduce both the nociceptive response and calcium levels.

7. Subdued RNAi increases the nociceptive response and C4da calcium levels in second instar larvae. Does overexpression of Subdued reduce the nociceptive response and C4da calcium levels in third instar larvae? This experiment would nicely show that Subdued levels are required for this developmental transition.

8. UAS controls are missing in figures 3 and 4.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled “Steroid hormone signaling activates thermal nociception during Drosophila peripheral nervous system development” for further consideration by eLife. Your revised article has been evaluated by K VijayRaghavan (Senior Editor) and a Reviewing Editor.

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

Essential revisions:

All revisions requested can be addressed by appropriate changes to the text.

In a revised discussion, please acknowledge and mention outstanding issues that need to be addressed by future physiological or calcium imaging experiments. Also, please make appropriate revisions in response to all of the itemised requests from Reviewer 3.

Reviewer #1:

The authors have improved the manuscript with their additional data and revisions. They have addressed most of my comments. The one issue that remains is the removal of the data for calcium levels at different developmental stages in response to temperature. The data in the manuscript shows calcium responses during periods of UV treatment are reduced in third instar larvae compared to second instar larvae. Subdued levels are reduced in third instar larvae and knockdown of subdued reduces calcium responses in second instar larvae. The question that remains is what happens to calcium levels in response to temperature at these developmental stages. I think these data are important in understanding how this developmental switch and subdued affects thermal nociception.

Reviewer #2:

This revision by the Jan group has addressed many of the issues that were raised in the first round of review. Specifically, staging of larvae has been improved especially with respect to the analysis of EcR immunostaining. Similarly, the text has been revised to address nuances related to EcR isoforms and ecdysone action across the L2-L3 larval stages. Finally, knockdown controls have been provided for RNAi transgenes, bolstering the conclusions. The removal of the calcium-imaging data for technical reasons is a little bit of a shame but the basic conclusions still stand on the improved behavioral and transcriptional analysis. The finding that EcRA and subdued are regulators of the L2-L3 transition in nociceptive sensitivity will be of interest to the fly field and those researchers more generally interested in mechanisms of modulation of nociceptive sensitivity.

Reviewer #3:

The resubmission of this paper is greatly improved over the original. I only have a few problems with the current version.

The first is with the Figures: I do not understand their philosophy for selecting data for the supplemental figures. The paper does not have many figures and the ones that they have are relatively simple. I do not see why important data are moved into supplemental figures. Figure 3 is a case in point. Part 3B presents third instar data showing that the W650A dominant negative (DN), suppresses the response to a 42C probe while the F645A-DN does not. The response latency data for this temperature are then shown in a supplemental Figure [lacking data for F645A-DN]. The supplemental Figure also shows the important data for the 46C probe showing both % response (for W650A-DN and F645A-DN) and latency (this also lacking F645A-DN data). It would be more helpful to the reader to have them combined in a single Figure. The same could be said for combining Figure 4 with Figure 4 Supp 1A-C.

In the body of Figures 3 through 5 they represent the isoforms with a period between EcR and the isoform designation (i.e. as “EcR.A” rather than “EcR-A”). They use the correct designations in the text and the figure legends (Also in Figure 4, Supp 1E-G). They should change the Figures to have a dash between EcR and its isoform designation.

Figure 2, Supp 1 A could confuse a reader into thinking that there are actually four EcR isoforms, the last being EcR-C. The last is a truncated, experimental construct. I suggest that the authors give it a name that reflects this – perhaps δ-EcR (using the Greek letter). This designation should be used throughout the text and in Figure 4A as well as Figure 2.

Figure 5: and supplement: I assume that in ppk-GAL4/EcR.A is actually ppk-GAL4/UAS-EcR.A.

The second problem is with the use of AF1 domains when one should be talking about A/B domains. In the Discussion the authors nicely come to grips with the complexity of ecdysone signaling and the diversity of its receptors. In the results, though, some of their Interim conclusions become confusing, especially with their usage of "AF1" rather than“"A/”" domain. AF1 refers to a functionality while A/B refers to the variable, N-terminal region of the receptor. If the terms are not used properly it can be confusing because EcR-A has an A/B domain but it has no AF1 function (see Hu et al., 2003). I suggest the following changes

Line 195: This sentence would be more clear as:“"the binding of ecdysone to EcR causes the recruitment of co-activators and the loss of co-repressors”"

Line 200 better to call it the“"co-activator recruitment domai”".

Line 204: F645A does block activation, but since it allows hormone binding, ligand-dependent de-repression can occur. W650A should prevent both de-repression and activation.

Line 219. Is this really correct that ecdysone“"activatio”" of EcR is required? Your data show that with the F645A-DN (which blocks ligand-dependent activation) show the appropriate shift in nociception. All of your data suggest that ecdysone is acting via de-repression

Line 233 More properly:“"difference between EcR isoforms is within their A/B domain”".

Line 235: should use Δ-EcR rather than EcR-C; same for ln 237, and 240.

Line 236: … to all isoforms and no A/B region.

Line 241-42: this is where it becomes confusing because EcR-A does not have a AF1 domain! The fact that EcR-A is effective while δ-EcR is not suggests that the A/B region of EcR-A carries a unique repressor function. Ecdysone binding is required to lose this repression (hence the differences between W650A and F645A). The paper gets around to this idea in the Discussion but the logic through the results is a bit muddy.

Line 277: better as …the EcR-A A/B domain and ligand binding are …..

Line 392 …. The unique A/B domain of the EcR-A isoform …..

The third issue that the paper focuses on the role of the isoforms so there should at least be images of EcR-A immunostaining along with EcR-common in Figure 2.

eLife. 2022 Mar 30;11:e76464. doi: 10.7554/eLife.76464.sa2

Author response


[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1:

This study by the Jan group examines the onset/development of functional thermal (and UV) nociception in Drosophila larvae. Previously, other groups had established that there are differences in nociceptive responses between second and third instar larvae (Cox) and that ecdysone receptor function is required for functional nociceptor development (Ganter). This study does a more thorough job than that previous work in documenting the stage differences and identifies a molecular player (subdued) that appears to be required for stage-specific changes in sensitivity. Overall, the data are sound and convincing and the paper is very well written. A few suggestions for improvement/clarification are offered.

1. Data organization. Some of the best data showing the stage-specific difference is in supplemental figure 1C- latency. This should really be in the main figure 1- probably as panel B. Figure 1C could benefit and would be more complete if second instar data were shown for 25 and 36 degrees.

We moved the latency data from supplemental figure 1C to the main figure 1 as Figure 1B. The original thermo-calcium data of Figure 1C has been removed due to technical issues that reduced our confidence in comparing calcium levels between different developmental stages at different temperatures.

2. The baseline calcium levels in second versus third instar is quite different- much higher in second instar looking at figure 1B. This data is essentially normalized/compared to same stage's baseline but this does raise the question: Does EcR function or subdued function shift the magnitude of response or shift the second instar (or third instar) baseline? It's a little hard to tell because Figure 3 (EcR) doesn't have stills of representative cells and Figure 6 (subdued, temperature) does not either. All we see here is the ratio data. Representative still data should be presented and whether the gene affects baseline calcium versus magnitude of change upon stimulus should be discussed.

Due to technical issues that reduced our confidence in comparing calcium levels between different developmental stages at different temperatures, we decided to remove the thermo-calcium data.

3. Figure 2 organization: Personal preference but the 20E data currently in supplement 2B is probably more important to the main point than the third instar data (Main figure 2D).

We moved 20E data from supplemental 2B to the main figure 3 as Figure 3A. Additionally, to better discuss the importance of the third instar EcR data, we have reformatted Figure 3 and moved the third instar dominant negative data to Figure 3B.

4. Some of the discussion of the differences between second and third instar seems a bit oversimplified, especially as it refers to the EcR expression data in Figure 2 supplement. It's not really accurate to say that second instar has low EcR and third has high. Reality (from the data shown) is that nuclear EcR is high in early second and later third and low in late second and early third. There isn't really data to address intra-stage differences in responsiveness so this should be described a bit more clearly- without the current oversimplification. Adding modifiers (late second instar, early third instar) would help.

To better examine the expression of EcR in second and third instar we have performed new IHC staining with improved developmental staging and quantification of EcR presence in the nucleus and cytoplasm. We have presented this data in a newly formatted Figure 2. We find that earlier in larval development, EcR is evenly distributed between the nucleus and cytoplasm in the second instar and at the beginning of the third instar. Nuclear localization increased 8 hrs after larval ecdysis and in wandering third instar larvae.

5. Analysis of subdued: Most helpful would be confirmation that this RNAi line is actually reducing subdued transcript levels, as one would predict. This would enhance confidence that there are no off-target effects. Also helpful would be the converse experiment to RNAi loss of function. Authors perform Subdued RNAi and see changes in second instar thermal responsiveness. Does overexpression of Subdued shift responsiveness in third instar?

We have now measured the ability of Subdued RNAi to reduce transcript levels by qRT-PCR and included the data in Figure 6 – supplemental figure 1C.

Overexpression of Subdued in C4da neurons has previously been reported by Jang et al. 2015, and the responsiveness of third instar larvae did not change. However, when Jang et al. 2015 used a subduedGal4 to drive Subdued overexpression, the number of fast responding third instar larvae increased. We have added discussion of these data to our Discussion section (Lines 437-443).

Reviewer #2:

The manuscript deals with the basis of an intriguing shift in nociceptive behavior shown by Drosophila larvae as they move from the L2 to the L3 instar. Prior publications have shown that the shift is controlled by ecdysone and implicated the EcR-A knock-down of EcR-A levels rendered the neurons hyposensitive to noxious thermal and mechanical stimulation and also altered features of their anatomy. Reduction in EcR-B1 did not effect their responses but has anatomical ramifications. This paper extends these studies into Ca2+ data from specific da neurons and also reveals a gene, subdued, that may play a role in the change in responsiveness.

Work by McParland et al. 2015 has shown that Ecdysone Receptor is required for nociception in the third instar, but it has remained unknown as to (A) what mechanism of EcR activity is required for nociception in the third instar (ligand-dependent or independent) and (B) whether ecdysone and EcR are involved in the developmental change of increased nociceptive behavior from early instars to the third instar.

There are a number of substantial concerns:

1. The paper focuses on the differences between the L2 and L3 instars. According to the Methods, their L2 stage is in the middle of the L2 period, but the L3 stage is a wandering L3 larva which is at least 50 hours older. There are three critical ecdysone-related events that occur in this 50 hour span: the critical weight at about 8-10 hr after L3 ecdysis, the mid-L3 transition, and the ecdysone peak that initiates wandering. If the transition occurs during the transition from L2 to L3, then the "critical weight" transition would be a good candidate and L3 larvae about 12 to 18 hr after ecdysis would be more appropriate.

To more precisely examine the expression of EcR in second and third instar we have performed new IHC staining with improved developmental staging and quantification of EcR presence in the nucleus and cytoplasm. We have presented this data in a newly formatted Figure 2. We find that earlier in larval development, EcR is evenly distributed between the nucleus and cytoplasm in the second instar and at the beginning of the third instar. Nuclear localization increased 8 hrs after L3 ecdysis and in wandering third instar larvae.

2. Another troubling methodological issue is that their behavioral tests were run at either 42 or 46 C while the Ca2+ measurements are done at 44 C. This becomes a concern in a situation like in Figure 2 where W650A expression is behaviorally effective at 42 C but has no effect at 46 C. We are then provided with a Ca2+ response at 44 C. Since this temp is half-way between the two used for the behavior tests, we have no way of assessing whether or not the Ca2+ response tracks the behavioral response.

We revised the manuscript to make a clear comparison of behavioral phenotypes with a 42oC probe as shown in the main figure: we have moved the 46oC data to supplement figures while expanding our explanation for preforming these experiments and our conclusions from these data. We have removed the calcium measurements after running into technical issues that reduced our confidence in comparing calcium levels between different developmental stages at different temperatures.

3. A serious interpretational issue has to do with ignoring the complexity imposed by the EcR isoforms. Through most of the paper and figures, the authors speak of “EcR” when they really should be saying “EcR-A”. This is not a trivial issue because the EcR-B isoforms have a strong hormone-independent transactivation domain at their N-terminus whereas EcR-A does not. At times when EcR-B1 is a present, the expression of wildtype EcR-A dampens the cellular response by competing with the endogenous EcR-B1 [Schubiger et al., 2003, Mech. Dev. 120:909]. Previous experiments (McPharland et al.) used RNAi to selectively knockdown different isoforms and these suggested a function for EcR-A. Ectopic expression of EcR-A alone, without EcR-B1 for comparison, can lead to wrong interpretations, especially since both isoforms are found in the da neurons (see McPharland).

We have addressed these concerns by: (1) Specifying which EcR isoform is being discussed, (2) Measuring EcR-A and EcR-B1 protein levels by using specific targeting antibodies in Figure 2, (3) Comparing the effects of overexpression of EcR-A, EcR-B1, and C on nociception in second and third instar larvae, and (4) Including an explanation of the interactions between EcR isoforms and interpretation of the overexpression experiments (Lines 232-246).

4. Another problem is with the interpretation of the EcR dominant negatives (EcR-DN). The function of nuclear receptors, like EcR, is too well understood to gloss over the impact of these point mutations by saying that they either prevent hormone binding [for W650A] or "co-factor" binding [for F645A]. The W650A mutation is in the binding pocket and prevents 20E binding. This prevents the loss of co-repressors and the assembly of a co-activator complex. Therefore, this forms serves as a constitutive repressor that is not responsive to 20E. The F645A substitution, by contrast is in the helix that changes conformation after 20E binding and is one of the residues involved in co-activator assembly. EcR-F645A allows 20E binding and the conformation changes that lose the co-repressors but it cannot support the co-activation. Therefore, it ca show hormone-dependent derepression but not activation.

We greatly appreciate the advice regarding how our writing about the activities of the EcR mutants can be improved. We have made revisions throughout the entire manuscript, including the explanation for the activity of the function of nuclear receptors (Lines 98-107) and the effects of the point mutations on their function (Lines 196-212).

5. Paragraph 299: This paragraph argues that expression of EcR[-A] in the L2 suppresses subdued expression and they conclude that expression of EcR-A in the L3 will enhance subdued expression because the EcR-A w650A suppresses subdued expression in the latter. They really need to show the effects of expression of the wild-type EcR-A receptor in the L3 as well as W650A. This information is important because as noted above, ectopic expression of EcR-A could act to suppress expression although likely not to the level seen for the W650A dominant negative.

We have tested the effect of expression of wild-type EcR-A expression on third instar nociception and found it neither reduced nor increased nociception. We have included this data in Figure 4 —figure supplement 2.

6. The interpretation of the W650A data in the L3 is confusing because its expression in the third suppresses both TrpA1 and subdued [Figure 4C] but has minimal effect on Ca2+ responses to heat [Figure 3] and has no effect on behavioral responses to 46 C [Figure 2D]. I was hoping that the Discussion would pull things together into a consistent story, but the Discussion tended to invoke a number of ad hoc hypotheses to account for pieces of data which do not seem to fit.

For the transcriptional data, upon further analysis of additional controls and nociceptive genes, as well as comparison to expression of EcR-W650A in second instar and third instar expression of EcR-A RNAi, we have found a broad effect of EcR-W650A on nociceptor transcription (Figure 5 —figure supplement 12) (we have moved the 46oC data to supplement figures.). Due to the broad effect of EcR-W650A on nociceptor transcription in second and third instar, and the limited effect of EcR-RNAi, we have narrowed our hypotheses in the discussion to the requirement of EcR-A ligand activity for maintaining appropriate expression of nociceptive genes, and the presence of EcR-A ligand-independent pathways which promote nociception in the third instar.

Reviewer #3:

The manuscript by Jaszczak et al. examines how steroid hormone signaling in Drosophila activates in sensory switch in C4da neurons that promotes a thermal nociceptive escape response behavior. The authors show that a greater percent of third instar larvae are able to generate the thermal nociceptive escape response than second instar larvae. They also show calcium responses in the soma of C4da neurons are larger in third instar larvae in response to high temperatures in comparison to second instar larvae. In contrast, they see the opposite with UV light. Both the behavioral response and the calcium response can be increased in second instar larvae by increasing ecdysone signaling. The authors also find that subdued expression is suppressed by increasing ecdysone signaling and link subdued to this developmental transition. Overall, the findings in the manuscript are interesting, but I think the authors need to address several points to strengthen their conclusions.

1. Some explanation as to why the soma is being imaged instead of dendrites or axons is needed. I assume the size of dendrites and axons make it too difficult to image in intact larvae.

Due to technical issues that reduced our confidence in comparing calcium levels in different developmental stages at different temperatures, we decided to remove the thermo-calcium data.

2. The authors mention that other groups have shown that EcR regulates dendritic growth. Could the role of EcR in this sensory switch be due to changes in dendritic growth instead of changes in calcium levels. This possibility should at least be discussed.

Measurement of dendrite architecture in second instar larvae overexpressing EcR-A has been added to the text and to Figure 4—figure supplement 1.

3. The authors use a dominant-negative approach (EcR-W650A) to reduce ecdysone signaling in third instar larvae. Dominant-negative approaches are difficult to interpret. The authors should use an EcR RNAi to knockdown ECR in C4da neurons. This should be done for both the thermal nociceptive escape responses and for the calcium imaging experiments.

By adding new data beyond what has been previously reported with EcR-RNAi (McParland et al. 2015) and comparing the effects of EcR-DN with those of EcR-RNAi, we now show nociceptor gene expression with expression of either EcR-DN or EcR-A RNAi in Figure 5 —figure supplement 2. While we did not find detectable transcriptional changes in nociceptor genes in third instar C4da neurons with EcR-A RNAi knockdown, we found broad suppression of nociceptor genes in third instar C4da neurons with EcR ligand binding mutant (W650A) expression (Figure 5 —figure supplement 2). We have incorporated this data into our discussion of the hypothesis that EcR-A ligand activity may function in a de-repressive capacity (Lines 383-407).

4. Why does the approach dominant-negative (EcR-W650A) reduce subdued expression in third instar larvae? I would expect that reduced ecdysone signaling would enhance subdued expression.

In order to further examine this effect of EcR-DN, we have compared nociceptor gene expression with EcR-DN at second and third instars (Figure 5 —figure supplement 2) and found a broad reduction of nociceptor gene expression at both stages of development. We have added to the Discussion the hypothesis that ligand dependent de-repression may be required for appropriate regulation of nociceptor gene transcription.

5. TrpA1 levels in second instar larvae are not affected by increased ecdysone signaling (overexpression of EcR) in second instar larvae, but they are reduced by expression of a dominant-negative (EcR-W650A). Again, this experiment should be repeated with an EcR RNAi for third instar larvae.

We have compared nociceptor gene expression with expression of either EcR-DN or EcR-A RNAi (Figure 5 —figure supplement 2). While we did not find detectable transcriptional changes in nociceptor genes in third instar C4da neurons with EcR-A RNAi expression, we found broad suppression of nociceptor genes in third instar C4da neurons with EcR ligand binding mutant (W650A) expression (Figure 5 —figure supplement 2).

6. If TrpA1 levels are regulated by ecdysone, does reducing TrpA1 in C4da neurons in third instar larvae reduce both the nociceptive response and calcium levels.

Upon further analysis of additional controls and nociceptive genes, as well as comparison to expression of EcR-W650A in second instar and third instar, we have found EcR-DN does not reduce TrpA1 in third instar larvae, while EcR-W650A has a broad effect on nociceptor transcription (Figure 5 —figure supplement 1-2). Reducing TrpA1 expression in third instar larvae has previously been shown to reduce nociception (Zhong et al. 2012; Terada et al. 2016; Gu et al. 2019) – we have referred to these previous studies in the revised manuscript.

7. Subdued RNAi increases the nociceptive response and C4da calcium levels in second instar larvae. Does overexpression of Subdued reduce the nociceptive response and C4da calcium levels in third instar larvae? This experiment would nicely show that Subdued levels are required for this developmental transition.

Overexpression of Subdued in C4da neurons has previously been reported by Jang et al. 2015 to yield no effect on the responsiveness of third instar larvae. However, when we used a subdued-Gal4 to drive Subdued overexpression, the number of fast responding third instar larvae increased. We have added these data to our discussion (Lines 437-443).

8. UAS controls are missing in figures 3 and 4.

Controls have been added for qRT-PCR experiments in Figure 5.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Essential revisions:

Reviewer #3:

The resubmission of this paper is greatly improved over the original. I only have a few problems with the current version.

The first is with the Figures: I do not understand their philosophy for selecting data for the supplemental figures. The paper does not have many figures and the ones that they have are relatively simple. I do not see why important data are moved into supplemental figures. Figure 3 is a case in point. Part 3B presents third instar data showing that the W650A dominant negative (DN), suppresses the response to a 42C probe while the F645A-DN does not. The response latency data for this temperature are then shown in a supplemental Figure [lacking data for F645A-DN]. The supplemental Figure also shows the important data for the 46C probe showing both % response (for W650A-DN and F645A-DN) and latency (this also lacking F645A-DN data). It would be more helpful to the reader to have them combined in a single Figure. The same could be said for combining Figure 4 with Figure 4 Supp 1A-C.

We have combined the Figure 3 and the supplement into a single figure. We have also added the latency data for the F645A genotypes. We have combined Figure 4 and Figure 4 —figure supplement 1 A-C into the primary Figure 4. We have also added the latency data for all of the EcR isoforms and the F645A genotypes and reference to data in the text (Line 213).

In the body of Figures 3 through 5 they represent the isoforms with a period between EcR and the isoform designation (i.e. as "EcR.A" rather than "EcR-A"). They use the correct designations in the text and the figure legends (Also in Figure 4, Supp 1E-G). They should change the Figures to have a dash between EcR and its isoform designation.

Figures 3, 4 , 4—figure supplement 1, 4—figure supplement 2, 5, and 5—figure supplement 1 have been corrected to change “.” to “-“ in EcR isoform labels.

Figure 2, Supp 1 A could confuse a reader into thinking that there are actually four EcR isoforms, the last being EcR-C. The last is a truncated, experimental construct. I suggest that the authors give it a name that reflects this – perhaps δ-EcR (using the Greek letter). This designation should be used throughout the text and in Figure 4A as well as Figure 2.

The label for the synthetic isoform has been changed to “EcR-ΔC” in the text and Figure 2 —figure supplement 1 and Figure 4. Additional designation as a “synthetic isoform” has also been added to the text and figure legend.

Figure 5: and supplement: I assume that in ppk-GAL4/EcR.A is actually ppk-GAL4/UAS-EcR.A.

“UAS-“ has been added to the transgene labels in figure 5 and Figure 5 —figure supplement 1.

The second problem is with the use of AF1 domains when one should be talking about A/B domains. In the Discussion the authors nicely come to grips with the complexity of ecdysone signaling and the diversity of its receptors. In the results, though, some of their interim conclusions become confusing, especially with their usage of "AF1" rather than "A/B" domain. AF1 refers to a functionality while A/B refers to the variable, N-terminal region of the receptor. If the terms are not used properly it can be confusing because EcR-A has an A/B domain but it has no AF1 function (see Hu et al., 2003). I suggest the following changes

Line 195: This sentence would be more clear as: "the binding of ecdysone to EcR causes the recruitment of co-activators and the loss of co-repressors."

Sentence changed as advised.

Line 200 better to call it the "co-activator recruitment domain".

Sentence changed as advised.

Line 204: F645A does block activation, but since it allows hormone binding, ligand-dependent de-repression can occur. W650A should prevent both de-repression and activation.

Line 200:206 changed to clarify that “W650A mutation prevents ligand binding and disrupts both derepression and activation. The F645A mutation cannot mediate activation, but retains the ligand binding capacity”. Line 202:204 has been added to clarify that ”…differences between the effect of overexpression of these mutant EcR-A constructs are likely due to differences in ligand binding ability”.

Line 219. Is this really correct that ecdysone "activation" of EcR is required? Your data show that with the F645A-DN (which blocks ligand-dependent activation) show the appropriate shift in nociception. All of your data suggest that ecdysone is acting via de-repression

Line 226 changed to clarify that “ligand activity through EcR-A is required…”.

Line 233 More properly: "difference between EcR isoforms is within their A/B domains".

Sentence changed as advised.

Line 235: should use Δ-EcR rather than EcR-C; same for ln 237, and 240.

EcR-ΔC is now used throughout the text.

Line 236: … to all isoforms and no A/B region.

Wording changed as advised.

Line 241-42: this is where it becomes confusing because EcR-A does not have a AF1 domain! The fact that EcR-A is effective while δ-EcR is not suggests that the A/B region of EcR-A carries a unique repressor function. Ecdysone binding is required to lose this repression (hence the differences between W650A and F645A). The paper gets around to this idea in the Discussion but the logic through the results is a bit muddy.

Line 277: better as …the EcR-A A/B domain and ligand binding are …..

Wording changed as advised.

Line 392 …. The unique A/B domain of the EcR-A isoform …..

Wording changed as advised.

The third issue that the paper focuses on the role of the isoforms so there should at least be images of EcR-A immunostaining along with EcR-common in Figure 2.

Images of EcR-A and EcR-B1 staining have been added to Figure 2.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    Transparent reporting form

    Data Availability Statement

    All data generated or analyzed during this study are included in the manuscript and supporting files.


    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES