Skip to main content
PMC home page
eLife logoLink to eLife
. 2016 Apr 29;5:e13254. doi: 10.7554/eLife.13254

The Ionotropic Receptors IR21a and IR25a mediate cool sensing in Drosophila

Lina Ni 1,2,3,†, Mason Klein 4,5,6,*,†, Kathryn V Svec 1,2,3, Gonzalo Budelli 1,2,3, Elaine C Chang 1,2,3, Anggie J Ferrer 5, Richard Benton 7, Aravinthan DT Samuel 4,6,*, Paul A Garrity 1,2,3,*
Editor: Ronald L Calabrese8
PMCID: PMC4851551  PMID: 27126188

Abstract

Animals rely on highly sensitive thermoreceptors to seek out optimal temperatures, but the molecular mechanisms of thermosensing are not well understood. The Dorsal Organ Cool Cells (DOCCs) of the Drosophila larva are a set of exceptionally thermosensitive neurons critical for larval cool avoidance. Here, we show that DOCC cool-sensing is mediated by Ionotropic Receptors (IRs), a family of sensory receptors widely studied in invertebrate chemical sensing. We find that two IRs, IR21a and IR25a, are required to mediate DOCC responses to cooling and are required for cool avoidance behavior. Furthermore, we find that ectopic expression of IR21a can confer cool-responsiveness in an Ir25a-dependent manner, suggesting an instructive role for IR21a in thermosensing. Together, these data show that IR family receptors can function together to mediate thermosensation of exquisite sensitivity.

DOI: http://dx.doi.org/10.7554/eLife.13254.001

Research Organism: D. melanogaster

eLife digest

Animals need to be able to sense temperatures for a number of reasons. For example, this ability allows animals to avoid conditions that are either too hot or too cold, and to maintain an optimal body temperature. Most animals detect temperature via nerve cells called thermoreceptors. These sensors are often extremely sensitive and some can even detect changes in temperature of just a few thousandths of a degree per second. However, it is not clear how thermoreceptors detect temperature with such sensitivity, and many of the key molecules involved in this ability are unknown.

In 2015, researchers discovered a class of highly sensitive nerve cells that allow fruit fly larvae to navigate away from unfavorably cool temperatures. Now, Ni, Klein et al. – who include some of the researchers involved in the 2015 work – have determined that these nerves use a combination of two receptors to detect cooling. Unexpectedly, these two receptors – Ionotropic Receptors called IR21a and IR25a – had previously been implicated in the detection of chemicals rather than temperature. IR25a was well-known to combine with other related receptors to detect an array of tastes and smells, while IR21a was thought to act in a similar way but had not been associated with detecting any specific chemicals. These findings demonstrate that the combination of IR21a and IR25a detects temperature instead.

Together, these findings reveal a new molecular mechanism that underlies an animal’s ability to sense temperature. These findings also raise the possibility that other “orphan” Ionotropic Receptors, which have not been shown to detect any specific chemicals, might actually contribute to sensing temperature instead. Further work will explore this possibility and attempt to uncover precisely how IR21a and IR25a work to detect cool temperatures.

DOI: http://dx.doi.org/10.7554/eLife.13254.002

Introduction

Temperature is an omnipresent physical variable with a dramatic impact on all aspects of biochemistry and physiology (Sengupta and Garrity, 2013). To cope with the unavoidable spatial and temporal variations in temperature they encounter, animals rely on thermosensory systems of exceptional sensitivity. These systems are used to avoid harmful thermal extremes and to seek out and maintain body temperatures optimal for performance, survival and reproduction (Barbagallo and Garrity, 2015; Flouris, 2011).

Among the most sensitive biological thermoreceptors described to date are the Dorsal Organ Cool Cells (DOCCs), a recently discovered trio of cool-responsive neurons found in each of the two dorsal organs at the anterior of the Drosophila melanogaster larva (Klein et al., 2015). The DOCCs robustly respond to decreases in temperature as small as a few millidegrees C per second (Klein et al., 2015), a thermosensitivity similar to that of the rattlesnake pit organ (Goris, 2011), a structure known for its extraordinary sensitivity. A combination of laser ablation, calcium imaging and cell-specific inhibition studies was used to establish the DOCCs as critical for mediating larval avoidance of temperatures below ~20˚C, with the thermosensitivity of this avoidance behavior paralleling the thermosensitivity of DOCC physiology (Klein et al., 2015). While the DOCCs are exceptionally responsive to temperature, the molecular mechanisms that underlie their thermosensitivity are unknown.

At the molecular level, several classes of transmembrane receptors have been shown to participate in thermosensation in animals. The most extensively studied are the highly thermosensitive members of the Transient Receptor Potential (TRP) family of cation channels (Palkar et al., 2015; Vriens et al., 2014). These TRPs function as temperature-activated ion channels and mediate many aspects of thermosensing from fruit flies to humans (Barbagallo and Garrity, 2015; Palkar et al., 2015; Vriens et al., 2014). In addition to TRPs, other classes of channels contribute to thermosensation in vertebrates, including the thermosensitive calcium-activated chloride channel Anoctamin 1 (Cho et al., 2012) and the two pore domain potassium channel TREK-1 (Alloui et al., 2006). Recent work in Drosophila has demonstrated that sensory receptors normally associated with other modalities, such as chemical sensing, can also make important contributions to thermotransduction. In particular, GR28B(D), a member of the invertebrate gustatory receptor (GR) family, was shown to function as a warmth receptor to mediate warmth avoidance in adult flies exposed to a steep thermal gradient (Ni et al., 2013). The photoreceptor Rhodopsin has also been reported to contribute to temperature responses, although its role in thermosensory neurons is unexamined (Shen et al., 2011).

Ionotropic Receptors (IRs) are a family of invertebrate receptors that have been widely studied in insect chemosensation, frequently serving as receptors for diverse acids and amines (Benton et al., 2009; Silbering et al., 2011). The IRs belong to the ionotropic glutamate receptor (iGluR) family, an evolutionarily conserved family of heterotetrameric cation channels that includes critical regulators of synaptic transmission like the NMDA and AMPA receptors (Croset et al., 2010). In contrast to iGluRs, IRs have been found only in Protostomia and are implicated in sensory transduction rather than synaptic transmission (Rytz et al., 2013). In insects, the IR family has undergone significant expansion and diversification, with the fruit fly D. melanogaster genome encoding 66 IRs (Croset et al., 2010). While the detailed structures of IR complexes are unknown, at least some IRs are thought to form heteromeric channels in which a broadly-expressed IR 'co-receptor' (such as IR25a, IR8a or IR76b) partners one or more selectively-expressed 'stimulus-specific' IRs (Abuin et al., 2011).

Among insect IRs, IR25a is the most highly conserved across species (Croset et al., 2010). In Drosophila, IR25a expression has been observed in multiple classes of chemosensory neurons with diverse chemical specificities, and IR25a has been shown to function as a 'co-receptor' that forms chemoreceptors of diverse specificities in combination with other, stimulus-specific IRs (Abuin et al., 2011; Rytz et al., 2013). IR21a is conserved in mosquitoes and other insects, but has not been associated with a specific chemoreceptor function (Silbering et al., 2011), raising the possibility that it may contribute to other sensory modalities.

Here, we show that the previously 'orphan' IR, Ir21a, acts together with the co-receptor IR25a to mediate thermotransduction. We show that these receptors are required for larval cool avoidance behavior as well as the physiological responsiveness of the DOCC thermosensory neurons to cooling. Furthermore, we find that ectopic expression of IR21a can confer cool responsiveness in an Ir25a-dependent manner, indicating that IR21a can influence thermotransduction in an instructive fashion.

Results

Dorsal organ cool cells express Ir21a-Gal4

To identify potential regulators of DOCC thermosensitivity, we sought sensory receptors specifically expressed in the dorsal organ housing these thermoreceptors (Figure 1a). Examining a range of potential sensory receptors in the larva, we found that regulatory sequences from the Ionotropic Receptor Ir21a drove robust gene expression (via the Gal4/UAS system [Brand and Perrimon, 1993]) in a subset of neurons in the dorsal organ ganglion, as well as in other locations (Figure 1b,c, Figure 1—figure supplement 1). Within each dorsal organ ganglion, Ir21a-Gal4 drove gene expression in three neurons (Figure 1b,c). These neurons exhibited the characteristic morphology of the DOCCs, which have unusual sensory processes that form a characteristic 'dendritic bulb' inside the larva (Klein et al., 2015).

Figure 1. Dorsal Organ Cool Cells (DOCCs) express Ir21a-Gal4.

(a) First/second instar larval anterior. Each Dorsal Organ Ganglion (grey) contains three DOCCs (blue). Anterior-Posterior axis denoted by double-headed arrow. (b,c) Ir21a-Gal4;UAS-GFP (Ir21a>GFP) labels larval DOCCs. Arrows denote cell bodies and arrowheads dendritic bulbs. (d) Temperature responses of Ir21a-Gal4;UAS-GCaMP6m-labeled DOCCs. Left panels, raw images; right panels, colors reflect fluorescence intensity. Arrows denote cell bodies. (e) Fluorescence quantified as percent change in fluorescence intensity compared to minimum intensity. n=22 cells (from 6 animals). (f,g) Temperature-responses of Ir21a-Gal4;R11F02-Gal4;UAS-GCaMP6m-labeled DOCCs. n=26 (7). Scale bars, 10 microns. Traces, average +/- SEM. Figure 1—figure supplement 1 provides an example of the 3-D imaging stacks used for calcium imaging data acquisition.

DOI: http://dx.doi.org/10.7554/eLife.13254.003

Figure 1.

Figure 1—figure supplement 1. Larval-wide expression patterns of Ir21a-Gal4 and R11F02-Gal4.

Figure 1—figure supplement 1.

(a) Ir21a-Gal4; UAS-GFP expression. (b) R11F02-Gal4; UAS-GFP expression. In addition to expression in the Dorsal Organ, both Gal4s exhibit expression in ~100 cells in the brain and ventral ganglion, neurons along the larval body wall and in the tail. R11F02-Gal4 is also expressed by sensory neurons in the Terminal Organ.
DOI: http://dx.doi.org/10.7554/eLife.13254.004
Figure 1—figure supplement 2. Calcium-imaging data are obtained as a three-dimensional imaging stack.

Figure 1—figure supplement 2.

(a) Dimensions of imaging volume. DOCCs depicted in blue. (b) Maximum intensity projections used for visualizing fluorescence intensity. (c) Representative image of maximum intensity projections of Ir21a>GCaMP6m-labeled DOCCs. DOCC cell bodies remain within imaging field throughout.
DOI: http://dx.doi.org/10.7554/eLife.13254.005
Open in a new tab

To confirm that the Ir21a-Gal4-positive neurons were indeed cool-responsive, their thermosensitivity was tested by cell-specific expression of the genetically encoded calcium indicator GCaMP6m under Ir21a-Gal4 control. Consistent with previously characterized DOCC responses (Klein et al., 2015), when exposed to a sinusoidal temperature stimulus between ~14˚C and ~20˚C, GCaMP6m fluorescence in these neurons increased upon cooling and decreased upon warming (Figure 1d,e and Figure 1—figure supplement 2). The expression of Ir21a-Gal4 was also compared with that of R11F02-Gal4 (Figure 1—figure supplement 1), a promoter used in the initial characterization of the DOCCs (Klein et al., 2015). As expected, GCaMP6m expressed under the combined control of Ir21a-Gal4 and R11F02-Gal4 revealed their precise overlap in three cool-responsive neurons with DOCC morphology in the dorsal organ, further confirming the identification of the Ir21a-Gal4-expressing cells as the cool-responsive DOCCs (Figure 1f,g).

Ir21a mediates larval thermotaxis

To assess the potential importance of Ir21a in larval thermosensation, we tested the ability of animals to thermotax when Ir21a function has been eliminated. Two Ir21a alleles were generated, Ir21a123and Ir21a∆1. Ir21a123 deletes 23 nucleotides in the region encoding the first transmembrane domain of IR21a and creates a translational frameshift (Figure 2a). Ir21a∆1 is an ~11 kb deletion removing all except the last 192 nucleotides of the Ir21a open reading frame, including all transmembrane and ion pore sequences (Figure 2a). As the deletion in Ir21a∆1 could also disrupt the nearby chitin deacetylase 5 (cda5) gene (Figure 2—figure supplement 1), Ir21a-specific rescue experiments were performed to confirm all defects reflected the loss of Ir21a activity (see below).

Figure 2. Larval cool avoidance requires Ir21a and Ir25a.

(a) Sequence alterations in Ir21a and Ir25a alleles. Ir21a regulatory sequences present in Ir21a-Gal4 are denoted in green and regions encoding transmembrane domains (TMs) and pore region in red. Additional details provided in Figure 2—figure supplement 1. (b) Thermotaxis is quantified as navigational bias. Cool avoidance behavior was assessed by tracking larval trajectories on a ~0.36˚C/cm gradient extending from ~13.5˚C to ~21.5˚C, with a midpoint of ~17.5˚C. (c) Cool avoidance requires Ir21a and Ir25a. Ir21a>Ir21a denotes a wild type Ir21a transcript expressed under Ir21a-Gal4 control. {Ir21a+} and {Ir25a+} denote wild type genomic rescue transgenes. Letters denote statistically distinct categories (alpha=0.05; Tukey HSD). wild type, n=836 animals. Ir21a∆1, n=74. Ir21a∆1;Ir21a-Gal4, n=48. Ir21a∆1;UAS-Ir21a, n=10. Ir21a∆1;Ir21a>Ir21a, n= 88. Ir21a∆1/ Ir21a123, n=71; Ir21a∆1/ Ir21a123; {Ir21a+} n=70; Ir25a2, n =100. Ir25a2; {Ir25a+} n= 247. Additional mutant analyses provided in Figure 2—figure supplement 1.

DOI: http://dx.doi.org/10.7554/eLife.13254.006

Figure 2.

Figure 2—figure supplement 1. Structure of Ir21a locus and analysis of thermotaxis in Ir8a and Ir76b mutants.

Figure 2—figure supplement 1.

(a) Cda5/Ir21a genomic region, denoting positions of the FRT-containing transposon insertions used to generate Ir21a∆1 (PBc04017 and PBc02720), the sequences deleted in Ir21a∆1, the Ir21a sequences present in the UAS-Ir21a rescue construct and the sequences present in the {Ir21+} genomic rescue construct. Untranslated regions are in gray. (b) Larval thermotaxis of Ir8a and Ir76b mutants quantified as navigational bias. Neither Ir8a nor Ir76b is required for cool avoidance; Ir8a mutants show enhanced cool avoidance compared to wild type. Letters denote statistically distinct categories (alpha=0.05; Tukey HSD). wild type, n=836 animals. Ir8a, n=166; Ir76b1, n=96, Ir76b2, n= 100.
DOI: http://dx.doi.org/10.7554/eLife.13254.007
Open in a new tab

The loss of Ir21a function strongly disrupted larval thermotaxis. When exposed to a thermal gradient of ~0.36˚C/cm, ranging from ~13.5˚C to ~21.5˚C, Ir21a∆1null mutants as well as Ir21a123/ Ir21a∆1heterozygotes were unable to navigate away from cooler temperatures and toward warmer temperatures (Figure 2b,c). These defects could be rescued by expression of a wild-type Ir21a transcript under Ir21a-Gal4 control and by a wild-type Ir21a genomic transgene (Figure 2c). Taken together, these results are consistent with a critical role for Ir21a in larval thermotaxis.

Ir25a mediates larval thermotaxis and is expressed in DOCCs

As IRs commonly act in conjunction with 'co-receptor' IRs, we examined the possibility that larval thermotaxis involved such additional IRs. Animals homozygous for loss-of-function mutations in two previously reported IR co-receptors, Ir8a and Ir76b, exhibited robust avoidance of cool temperatures, indicating that these receptors are not essential for this behavior (Figure 2—figure supplement 1). By contrast, Ir25a2 null mutants failed to avoid cool temperatures, a defect that could be rescued by the introduction of a transgene containing a wild type copy of Ir25a (Figure 2c). Thus, Ir25a also participates in cool avoidance. To assess IR25a expression, larvae were stained with antisera for IR25a. Robust IR25a protein expression was detected in multiple cells in the dorsal organ ganglion, including the three Ir21a-Gal4-expressing DOCCs (Figure 3a). Within DOCCs, IR25a strongly labels the 'dendritic bulbs', consistent with a role in sensory transduction. Staining was absent in Ir25a null mutants demonstrating staining specificity (Figure 3b). Thus Ir25a is required for thermotaxis and is expressed in the neurons that drive this behavior.

Figure 3. DOCCs express IR25a.

Figure 3.

Open in a new tab

(a) Left panel, Ir21a>GFP-labeled DOCCs. Middle panel, IR25a protein expression in dorsal organ. Right panel, Ir21a>GFP-labeled DOCCs express IR25a protein. Arrows denote DOCC cell bodies and arrowheads DOCC dendritic bulbs. (b) IR25a immunostaining is not detected in Ir25a2null mutants. Scale bar, 10 microns.

DOI: http://dx.doi.org/10.7554/eLife.13254.008

Ir21a and Ir25a are required for cool detection by DOCCs

To assess whether Ir21a and Ir25a contribute to cool detection by the DOCCs, DOCC cool-responsiveness was examined using the genetically encoded calcium sensor GCaMP6m. Consistent with a role for Ir21a in cool responses, DOCCs exhibited strongly reduced responses to cooling in Ir21a∆1 deletion mutants, and this defect was robustly rescued by expression of an Ir21a transcript in the DOCCs using R11F02-Gal4 (Figure 4a-e,h). Similarly, DOCC thermosensory responses were greatly reduced in Ir25a mutants, a defect that was rescued by a wild type Ir25a transgene (Figure 4f–h). Together these data demonstrate a critical role for Ir21a and Ir25a in the detection of cooling by the DOCCs.

Figure 4. DOCC cool responses require Ir21a and Ir25a.

DOCC responses monitored using R11F02>GCaMP6m. DOCCs exhibit robust cool-responsive increases in fluorescence (a,c), which are dramatically reduced in Ir21a (b,d) and Ir25a (f) mutants. (e) Ir21a transcript expression under R11F02-Gal4 control rescues the Ir21a mutant defect. (g) Introduction of an Ir25a genomic rescue transgene rescues the Ir25a mutant defect. (h) Ratio of fluorescence at 14˚C versus 20˚C depicted using a violin plot. Letters denote statistically distinct categories, p<0.0001, Steel-Dwass test. Scale bars, 10 microns. Traces, average +/- SEM. wild type, n=33 cells (from 11 animals). Ir21a∆1, n= 58 (14). Ir21a∆1; R11F02>Ir21a, n=32 (9). Ir25a2, n=43 (16). Ir25a2; {Ir25a+}, n=30 (10). Analyses of brv1 and brv2 mutants provided in Figure 4—figure supplement 1.

DOI: http://dx.doi.org/10.7554/eLife.13254.009

Figure 4.

Figure 4—figure supplement 1. Analysis of putative null mutants of brv1 and brv2.

Figure 4—figure supplement 1.

(a) brv1 but not brv2 mutants exhibit defects in larval cool avoidance. Thermotaxis quantified as navigational bias. Letters denote statistically distinct categories (alpha=0.05; Tukey HSD). wild type, n=836 animals. brv1L653stop, n =43. brv2W205stop, n =99. b) Ir21a>GCaMP6m-labelled DOCCs respond to cooling in brv1L653stopmutants. n= 35 cells (from 6 animals).
DOI: http://dx.doi.org/10.7554/eLife.13254.010
Open in a new tab

Prior work has suggested that three TRP channels, Brivido-1, Brivido-2 and Brivido-3, work together to mediate cool sensing in adult thermosensors (Gallio et al., 2011). Putative null mutations are available for two of these genes, brv1 and brv2, and we used these alleles to test the potential role of Brivido function in DOCC cool sensing (Gallio et al., 2011). Although brv1 mutant showed defects in thermotactic behavior, DOCC responses to cooling appeared unaffected in brv1 mutants (Figure 4—figure supplement 1). brv2 nulls exhibited no detectable thermotaxis defects (Figure 4—figure supplement 1). Thus, we detect no role for these receptors in cool sensing by the DOCCs.

Ectopic IR21a expression confers cool-sensitivity in an Ir25a-dependent manner

The requirement for Ir21a and Ir25a in DOCC-mediated cool sensing raised the question of whether ectopic expression of these receptors could confer cool-responsiveness upon a cell, as might be predicted for a cool receptor. Attempts to express IR21a and IR25a together or separately in heterologous cells, including S2 cells, Xenopus oocytes and HEK cells, failed to yield detectable responses to cooling or warming, as did attempts to confer thermosensitivity upon non-thermosensitive neurons by ectopically expressing them separately or together in Drosophila, broadly throughout the larval nervous system and in adult chemosensory neurons (G.B., L.N., M.K. and P.G, unpublished). However, ectopic expression of IR21a in one set of neurons in the adult, Hot Cell thermoreceptors in the arista that normally respond to warming rather than cooling, conferred cool-sensitivity.

The adult arista contains three warmth-activated thermosensory neurons, termed Hot Cells (or HC neurons) (Gallio et al., 2011). We found that forced expression of IR21a in the HC neurons could significantly alter their response to temperature. As previously reported (Gallio et al., 2011), wild-type HC neurons respond to warming with robust increases in intracellular calcium and to cooling with decreases in intracellular calcium, as reflected in temperature-dependent changes in GCaMP6m fluorescence (Figure 5a,c). In contrast, HC neurons in which IR21a is expressed under the control of a pan-neuronal promoter (N-syb>Ir21a animals) frequently exhibited elevations in calcium not only in response to warming, but also at the coolest temperatures (Figure 5b,d,f, Figure 5—figure supplement 1a). Thus, ectopic IR21a expression causes HC neurons, which are normally inhibited by cooling, to become responsive to both cooling and warming.

Figure 5. IR21a expression confers cool-sensitivity upon warmth-responsive Hot Cell neurons.

(a,b) Temperature responses of wild type (a) or N-syb>Ir21a-expressing (b) thermoreceptors in the arista, monitored with N-syb>GCaMP6m. Cell bodies of warmth-responsive Hot Cells outlined in red and cool-responsive Cold Cells in blue. Arrows highlight Hot Cells at 14˚C. Traces of Hot Cell and Cold Cell responses shown at right. Scale bar, 10 microns. (c-e) Fluorescence of Hot Cells in response to sinusoidal 14˚C to 30˚C temperature stimulus, quantified as percent ∆F/Fmin. Dotted lines denote temperature minima. Traces, average +/- SEM. (f) Difference between ∆F/Fmin at 14˚C vs 20˚C (average +/- SEM). Responses of N-syb>Ir21a cells were statistically distinct from both wild type and Ir25a2;N-syb>Ir21a (p<0.01, Steel-Dwass test; letters denote statistically distinct groups). wild type, n= 16 cells (from 8 animals). N-syb>Ir21a, n= 16 (10). Ir25a2; N-syb>Ir21a, n= 20 (10). Analysis of endogenous IR25a expression in the Hot Cells and of the consequences of Hot Cell-specific misexpression of IR21a provided in Figure 5—figure supplement 1.

DOI: http://dx.doi.org/10.7554/eLife.13254.011

Figure 5.

Figure 5—figure supplement 1. Hot Cell neurons express IR25a protein, and IR21a confers cool-sensitivity upon the Hot Cell neurons.

Figure 5—figure supplement 1.

(a) Difference between ∆F/Fmin at 14˚C vs 20˚C for each cell imaged in Figure 5. (b, c) Left panel, HC>GFP-labeled Hot Cell neurons. Middle panel, IR25a immunostaining. Right panel, HC>GFP and IR25a co-expression. Arrows indicate Hot Cell neuron cell bodies. Specific IR25a immunostaining is absent in Ir25a null mutants (b). (d, e) Temperature responses of wild type (d), HC>Ir21a-expressing (e) thermoreceptors in the adult arista, monitored using HC>GCaMP6m. Dotted lines denote temperature minima. Traces, average +/- SEM. wild type, n=4 cells (2 animals). HC>IR21a n=25 (9).
DOI: http://dx.doi.org/10.7554/eLife.13254.012
Open in a new tab

As Ir21a-dependent cool detection in the DOCCs relies upon Ir25a, we examined the requirement for Ir25a in IR21a-mediated cool activation of the HC neurons. Consistent with previously reported IR25a expression in the arista (Benton et al., 2009), we observed robust IR25a protein expression in the HC neurons (Figure 5—figure supplement 1b,c). Consistent with a role for Ir25a in Ir21a-mediated cool-responsiveness, ectopic IR21a expression failed to drive significant HC neuron cool responses in Ir25a mutants (Figure 5e,f). Thus, IR21a can confer cool-sensitivity upon an otherwise warmth-responsive neuron in an Ir25a-dependent fashion. Similar cool sensitivity was observed when IR21a was ectopically expressed under the control of an HC-specific promoter (HC>Ir21a, Figure 5—figure supplement 1d,e). Finally, ectopic expression of IR21a in Gr28b mutant HC neurons, which lack the Gr28b(D) warmth receptor, yields neurons that respond only to cooling (Figure 6). Together, these data demonstrate that ectopic IR21a expression can confer cool-sensitivity in an Ir25a-dependent fashion.

Figure 6. Hot Cell-specific expression of IR21a confers cool-sensitivity upon Gr28b mutant Hot Cell neurons.

Figure 6.

Open in a new tab

(a-c) Temperature responses of wild type (a), Gr28b mutant (b), and HC>Ir21a-expressing Gr28b mutant (c) thermoreceptors in the adult arista, monitored using HC>GCaMP6m. Dotted lines denote temperature minima. Traces, mean +/- SEM. wild type, n=11 cells (3 animals). Gr28bMi n=9 (3). HC>IR21a; Gr28bMi n=11 (3). (d) Cool responses (∆F/F14˚C - ∆F/F30˚C) of HC>IR21a; Gr28bMicells were distinct from both wild type and Gr28bMi (p<0.01, Steel-Dwass test, letters denote statistically distinct groups).

DOI: http://dx.doi.org/10.7554/eLife.13254.013

Discussion

These data demonstrate that the ionotropic receptors IR21a and IR25a have critical roles in thermosensation in Drosophila, mediating cool detection by the larval dorsal organ cool cells (DOCCs) and the avoidance of cool temperatures. Combinations of IRs have been previously found to contribute to a wide range of chemosensory responses, including the detection of acids and amines (Rytz et al., 2013). These findings extend the range of sensory stimuli mediated by these receptor combinations to cool temperatures. Interestingly, IR21a- and IR25a-dependent cool sensation appears independent of Brivido 1 and Brivido 2, two TRP channels implicated in cool sensing in the adult (Gallio et al., 2011).

The precise nature of the molecular complexes that IRs form is not well understood. IR25a has been shown to act with other IRs in the formation of chemoreceptors, potentially as heteromers (Rytz et al., 2013). This precedent raises the appealing possibility that IR25a might form heteromeric thermoreceptors in combination with IR21a. However, our inability to readily reconstitute temperature-responsive receptor complexes in heterologous cells suggests that the mechanism by which these receptors contribute to cool responsiveness is likely to involve additional molecular cofactors. It is interesting to note that the range of cell types in which ectopic IR21a expression confers cool-sensitivity is so far restricted to neurons that already respond to temperature. This observation suggests the existence of additional co-factors or structures in these thermosensory cells that are critical for IR21a and IR25a to mediate responses to temperature. All studies to date implicate IRs as receptors for sensory stimuli (Rytz et al., 2013), and our misexpression studies are consistent with a similar role for Ir21a and IR25a in cool sensation. However, we cannot formally exclude the possibility that they could have indirect, and possibly separate, functions in this process, for example, in regulating the expression or function of an unidentified cool receptor. Interestingly, IR25a was recently implicated in warmth-responsive resetting of the circadian clock, and suggested to confer warmth-sensitivity on its own, without the co-expression of other IRs (Chen et al., 2015). The ability of IR25a to serve as a warmth receptor on its own would be a surprise given both its broad expression and its established role as an IR co-receptor (Abuin et al., 2011). As IR25a misexpression only slightly enhanced the thermosensitivity of an already warmth-responsive neuron (Chen et al., 2015), this raises the alternative possibility that – analogous to cool-sensing – IR25a acts not on its own, but rather as a co-receptor with other IRs involved in warmth-sensing.

While the present study focuses on the role of IR21a and IR25a in larval thermosensation, it is interesting to note that the expression of both IR21a and IR25a has been detected in the thermoreceptors of the adult arista (Benton et al., 2009). Thus, related mechanisms could contribute to thermosensory responses not only in the DOCCs, but also in other cellular contexts and life stages. Moreover, the presence of orthologs of IR21a and IR25a across a range of insects (Croset et al., 2010) raises the possibility that these IRs, along other members of the IR family, constitute a family of deeply-conserved thermosensors.

Materials and methods

Fly strains

Ir25a2(Benton et al., 2009>), BAC{Ir25a+} (Chen et al., 2015), Ir8a1(Abuin et al., 2011), Ir76b1 (Zhang et al., 2013), Ir76b2(Zhang et al., 2013), R11F02-Gal4 (Klein et al., 2015), brv1L653stop(Gallio et al., 2011), brv2w205stop (Gallio et al., 2011), HC-Gal4 (Gallio et al., 2011), Gr28bMi (Ni et al., 2013), UAS-GCaMP6m (P{20XUAS-IVS-GCaMP6m}attp2 and P{20XUAS-IVS-GCaMP6m}attp2attP40 [Chen et al., 2013]), UAS-GFP (p{10X UAS-IVS-Syn21-GFP-p10}attP2 [Pfeiffer et al., 2012]), nSyb-Gal4 (P{GMR57c10-Gal4}attP2, [Pfeiffer et al., 2012]), and y1 P(act5c-cas9, w+) M(3xP3-RFP.attP)ZH-2A w* (Port et al., 2014) were previously described.

In Ir21a-Gal4, sequences from -606 to +978 with respect to the Ir21a translational start site (chromosome 2L: 24,173 – 25757, reverse complement) lie upstream of Gal4 protein-coding sequences. UAS-Ir21a contains the Ir21a primary transcript including introns (chromosome 2L: 21823–25155, reverse complement) placed under UAS control. The {Ir21a+} genomic rescue construct contains sequences from -1002 to +4439 with respect to the Ir21a translational start site (chromosome 2L: 26153–20712).

Ir21a∆1 was generated by FLP-mediated recombination between two FRT-containing transposon insertions (PBac{PB}c02720 and PBac{PB}c04017) as described (Parks et al., 2004). Ir21a123 was generated by transgene-based CRISPR-mediated genome engineering as described (Port et al., 2014), with an Ir21a-targeting gRNA (5’-CTGATTTGCGTTTACCTCGG) expressed under U6-3 promoter control (dU6-3:gRNA) in the presence of act-cas9 (Port et al., 2014).

Behaviour

Thermotaxis of early 2nd instar larvae was assessed over a 15 min period on a temperature gradient extending from 13.5 to 21.5°C over 22 cm (~0.36˚C/cm) as described (Klein et al., 2015). As behavioral data appear normally distributed (as assessed by Shapiro-Wilk test), statistical comparisons were performed by Tukey HSD test, which corrects for multiple comparisons.

Calcium imaging

Calcium imaging was performed as previously described for larvae (Klein et al., 2015). Pseudocolor images were created using the 16_colors lookup table in ImageJ 1.43r. Adult calcium imaging was performed as described for larvae (Klein et al., 2015), with modifications to the temperature stimulus and sample preparation approach. Adult temperature stimulus ranged from 14°C to 30°C. Intact adult antennae with aristae attached were dissected and placed in fly saline (110 mM NaCl, 5.4 mM KCl, 1.9 mM CaCl2, 20 mM NaHCO3, 15 mM tris(hydroxymethyl)aminomethane (Tris), 13.9 mM glucose, 73.7 mM sucrose, and 23 mM fructose, pH 7.2, [Brotz and Borst, 1996]) on a large cover slip (24 mm x 50 mm) and then covered by a small cover slip (18 mm x 18 mm). The large cover slip was placed on top of a drop of glycerol on the temperature control stage. As quantified calcium imaging data (Figure 4h, Figure 5f, Figure 6d) did not conform to a normal distribution as assessed by Shapiro-Wilk test (p<0.01), statistical comparisons were performed by Steel-Dwass test, a non-parametric test that corrects for multiple comparisons, using JMP11 (SAS).

Immunohistochemistry

Immunostaining was performed as described (Kang et al., 2012) using rabbit anti-Ir25a (1:100; [Benton et al., 2009]), mouse anti-GFP (1:200; Roche), goat anti-rabbit Cy3 (1:100; Jackson ImmunoResearch), donkey anti-mouse FITC (1:100; Jackson ImmunoResearch).

Acknowledgements

We thank Rachelle Gaudet and Linda Huang for comments on the manuscript, Guangwei Si for assistance with calcium imaging, Peter Bronk for advice on physiology, Adam Kaplan for creating Ir21a-Gal4, and the Bloomington Stock Center for fly strains. Supported by a grant from the National Institute of Neurological Disorders and Stroke (F32 NS077835) to MK, the National Institute of General Medical Science (F32 GM113318) to GB, European Research Council Starting Independent Researcher and Consolidator Grants (205202 and 615094) to RB, and the National Institute of General Medical Sciences (P01 GM103770) to ADTS and PAG.

Funding Statement

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

Funding Information

This paper was supported by the following grants:

  • National Institute of Neurological Disorders and Stroke F32 NS077835 to Mason Klein.

  • National Institute of General Medical Sciences F32 GM113318 to Gonzalo Budelli.

  • European Research Council 205202 to Richard Benton.

  • European Research Council 615094 to Richard Benton.

  • National Institute of General Medical Sciences P01 GM103770 to Aravinthan DT Samuel, Paul A Garrity.

Additional information

Competing interests

The authors declare that no competing interests exist.

Author contributions

LN, Performed molecular genetics, behavior, immunohistochemistry and calcium imaging, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents.

MK, Performed behavior and calcium imaging, Performed data analysis, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents.

KVS, Performed molecular genetics, Acquisition of data, Contributed unpublished essential data or reagents.

GB, Performed physiology, Conception and design, Acquisition of data, Analysis and interpretation of data, Contributed unpublished essential data or reagents.

ECC, Performed molecular genetics, Acquisition of data, Contributed unpublished essential data or reagents.

AJF, Performed data analysis, Analysis and interpretation of data.

RB, Conception and design, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents.

ADTS, Conception and design, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents.

PAG, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents.

Ethics

Animal experimentation: This study (specifically the harvest of Xenopus laevis oocytes) was performed in strict accordance with approved institutional animal care and use committee (IACUC) protocol (#14077) of Brandeis University.

References

  1. Abuin L, Bargeton B, Ulbrich MH, Isacoff EY, Kellenberger S, Benton R. Functional architecture of olfactory ionotropic glutamate receptors. Neuron. 2011;69:44–60. doi: 10.1016/j.neuron.2010.11.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alloui A, Zimmermann K, Mamet J, Duprat F, Noël J, Chemin J, Guy N, Blondeau N, Voilley N, Rubat-Coudert C, Borsotto M, Romey G, Heurteaux C, Reeh P, Eschalier A, Lazdunski M. TREK-1, a K+ channel involved in polymodal pain perception. The EMBO Journal. 2006;25:2368–2376. doi: 10.1038/sj.emboj.7601116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barbagallo B, Garrity PA. Temperature sensation in Drosophila. Current Opinion in Neurobiology. 2015;34:8–13. doi: 10.1016/j.conb.2015.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Benton R, Vannice KS, Gomez-Diaz C, Vosshall LB. Variant ionotropic glutamate receptors as chemosensory receptors in Drosophila. Cell. 2009;136:149–162. doi: 10.1016/j.cell.2008.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brand AH, Perrimon N. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development. 1993;118:401–415. doi: 10.1242/dev.118.2.401. [DOI] [PubMed] [Google Scholar]
  6. Brotz TM, Borst A. Cholinergic and GABAergic receptors on fly tangential cells and their role in visual motion detection. Journal of Neurophysiology. 1996;76:1786–1799. doi: 10.1152/jn.1996.76.3.1786. [DOI] [PubMed] [Google Scholar]
  7. Chen TW, Wardill TJ, Sun Y, Pulver SR, Renninger SL, Baohan A, Schreiter ER, Kerr RA, Orger MB, Jayaraman V, Looger LL, Svoboda K, Kim DS. Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature. 2013;499:295–300. doi: 10.1038/nature12354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chen C, Buhl E, Xu M, Croset V, Rees JS, Lilley KS, Benton R, Hodge JJ, Stanewsky R. Drosophila Ionotropic Receptor 25a mediates circadian clock resetting by temperature. Nature. 2015;527:516–520. doi: 10.1038/nature16148. [DOI] [PubMed] [Google Scholar]
  9. Cho H, Yang YD, Lee J, Lee B, Kim T, Jang Y, Back SK, Na HS, Harfe BD, Wang F, Raouf R, Wood JN, Oh U. The calcium-activated chloride channel anoctamin 1 acts as a heat sensor in nociceptive neurons. Nature Neuroscience. 2012;15:1015–1021. doi: 10.1038/nn.3111. [DOI] [PubMed] [Google Scholar]
  10. Croset V, Rytz R, Cummins SF, Budd A, Brawand D, Kaessmann H, Gibson TJ, Benton R. Ancient protostome origin of chemosensory ionotropic glutamate receptors and the evolution of insect taste and olfaction. PLoS Genetics. 2010;6:e13254. doi: 10.1371/journal.pgen.1001064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Flouris AD. Functional architecture of behavioural thermoregulation. European Journal of Applied Physiology. 2011;111:e13254. doi: 10.1007/s00421-010-1602-8. [DOI] [PubMed] [Google Scholar]
  12. Gallio M, Ofstad TA, Macpherson LJ, Wang JW, Zuker CS. The coding of temperature in the Drosophila brain. Cell. 2011;144:614–624. doi: 10.1016/j.cell.2011.01.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Goris RC. Infrared organs of snakes: an integral part of vision. Journal of Herpetology. 2011;45:2–14. doi: 10.1670/10-238.1. [DOI] [Google Scholar]
  14. Kang K, Panzano VC, Chang EC, Ni L, Dainis AM, Jenkins AM, Regna K, Muskavitch MAT, Garrity PA. Modulation of TRPA1 thermal sensitivity enables sensory discrimination in Drosophila. Nature. 2012;481:76–80. doi: 10.1038/nature10715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Klein M, Afonso B, Vonner AJ, Hernandez-Nunez L, Berck M, Tabone CJ, Kane EA, Pieribone VA, Nitabach MN, Cardona A, Zlatic M, Sprecher SG, Gershow M, Garrity PA, Samuel AD. Sensory determinants of behavioral dynamics in Drosophila thermotaxis. Proceedings of the National Academy of Sciences of the United States of America. 2015;112:e13254. doi: 10.1073/pnas.1416212112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ni L, Bronk P, Chang EC, Lowell AM, Flam JO, Panzano VC, Theobald DL, Griffith LC, Garrity PA. A gustatory receptor paralogue controls rapid warmth avoidance in Drosophila. Nature. 2013;500:580–584. doi: 10.1038/nature12390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Palkar R, Lippoldt EK, McKemy DD. The molecular and cellular basis of thermosensation in mammals. Current Opinion in Neurobiology. 2015;34:14–19. doi: 10.1016/j.conb.2015.01.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Parks AL, Cook KR, Belvin M, Dompe NA, Fawcett R, Huppert K, Tan LR, Winter CG, Bogart KP, Deal JE, Deal-Herr ME, Grant D, Marcinko M, Miyazaki WY, Robertson S, Shaw KJ, Tabios M, Vysotskaia V, Zhao L, Andrade RS, Edgar KA, Howie E, Killpack K, Milash B, Norton A, Thao D, Whittaker K, Winner MA, Friedman L, Margolis J, Singer MA, Kopczynski C, Curtis D, Kaufman TC, Plowman GD, Duyk G, Francis-Lang HL. Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genome. Nature Genetics. 2004;36:288–292. doi: 10.1038/ng1312. [DOI] [PubMed] [Google Scholar]
  19. Pfeiffer BD, Truman JW, Rubin GM. Using translational enhancers to increase transgene expression in Drosophila. Proceedings of the National Academy of Sciences of the United States of America. 2012;109:6626–6631. doi: 10.1073/pnas.1204520109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Port F, Chen HM, Lee T, Bullock SL. Optimized CRISPR/Cas tools for efficient germline and somatic genome engineering in Drosophila. Proceedings of the National Academy of Sciences of the United States of America. 2014;111:e13254. doi: 10.1073/pnas.1405500111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Rytz R, Croset V, Benton R. Ionotropic receptors (IRs): chemosensory ionotropic glutamate receptors in Drosophila and beyond. Insect Biochemistry and Molecular Biology. 2013;43:888–897. doi: 10.1016/j.ibmb.2013.02.007. [DOI] [PubMed] [Google Scholar]
  22. Sengupta P, Garrity P. Sensing temperature. Current Biology. 2013;23:R304–307. doi: 10.1016/j.cub.2013.03.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Shen WL, Kwon Y, Adegbola AA, Luo J, Chess A, Montell C. Function of rhodopsin in temperature discrimination in Drosophila. Science. 2011;331:1333–1336. doi: 10.1126/science.1198904. [DOI] [PubMed] [Google Scholar]
  24. Silbering AF, Rytz R, Grosjean Y, Abuin L, Ramdya P, Jefferis GS, Benton R. Complementary function and integrated wiring of the evolutionarily distinct Drosophila olfactory subsystems. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience. 2011;31:13357–13375. doi: 10.1523/JNEUROSCI.2360-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Vriens J, Nilius B, Voets T. Peripheral thermosensation in mammals. Nature Reviews. Neuroscience. 2014;15:573–589. doi: 10.1038/nrn3784. [DOI] [PubMed] [Google Scholar]
  26. Zhang YV, Ni J, Montell C. The molecular basis for attractive salt-taste coding in Drosophila. Science. 2013;340:1334–1338. doi: 10.1126/science.1234133. [DOI] [PMC free article] [PubMed] [Google Scholar]
eLife. 2016 Apr 29;5:e13254. doi: 10.7554/eLife.13254.014

Decision letter

Editor: Ronald L Calabrese1

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your work entitled "The Ionotropic Receptors IR21a and IR25a mediate cool sensing in Drosophila" for consideration by eLife. Your article has been favorably evaluated by K VijayRaghavan (Senior editor) and three reviewers, one of whom, (Ronald L. Calabrese) is a member of our Board of Reviewing Editors.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

This research report presents convincing genetic, calcium imaging, and behavioral data that specific Ionotropic Receptors (IRs) are required for cool sensing in Dorsal Organ Cool Cells (DOCCs) of the Drosophila larva and for cool avoidance (thermotaxis) in larvae. IR21a and IR25a appear to work together to confer cool sensing to the DOCCs. Moreover, ectopic expression of IR21a in thermosensitive neurons confers a component of cool sensitivity but requires IR25a expression. The demonstration that temperature sensing can be mediated by IRs has wide implication for the study of thermostatic behavior throughout the Protostomia, particularly insects.

The experiments are carefully performed and appropriately analyzed with relevant statistics. The Figures are easy to assimilate and the legends clear. Supplementary data answers for important controls. Materials and methods is adequate. Writing is very clear.

Essential revisions:

There are some concerns however that must be addressed before the manuscript can be published in eLife. While it is clear that IR21a and IR25a are necessary for cool sensing in DOCCs, it is not completely convincing that they themselves are cool receptors. As one expert reviewer points out "In contrast to earlier work showing that ectopic expression of Drosophila TRPA1 or GR28B(D) confers heat sensitivity to various cell types (e.g. Ni et al. Nature 2013 from the same group), this does not seem to be the case for IR21a and IR25a, except for a small response in the already thermosensitive Hot Cell neurons. Therefore, various other plausible scenarios can be envisaged where deletion of these proteins would abolish a cold response, without them being involved in the actual cold sensing. For instance, they could act downstream of the actual cold sensor (e.g. similar to a voltage-gated calcium channel at a sensory nerve ending), or their genetic elimination could merely influence the expression of the actual cold sensor. Similarly, it is suggested that these proteins act in a molecular complex, possibly as heteromultimers, but there is no evidence presented that these IRs interact with each other at the molecular level. Given that IR25a has already recently been implicated in thermosensing (Chen et al. 2015), I feel that, for this paper to represent a sufficient advance to merit publication in eLife, additional evidence supporting the hypothesis of IR21a/IR25a complexes as cold sensor should be provided. Minimally the authors should do an experiment where IR21a is expressed ectopically in thermosensitive neurons lacking a hot response (ectopic expression in a Gr28 mutant) to show robust confirmation of cool sensing. Moreover, that authors should discuss their results in light of these caveats.

There are two other concerns from the expert reviewers that should be addressed.

1) What is the demonstration that the 3 sensory cells that respond to cool are responsible for the behavioral phenotypes? Complete expression patterns of Ir21a-Gal4, Ir25a-Gal4, and R11F02 are necessary to evaluate this. If Ir21a or R11F02 is only in the 3 cool cells, this is easily resolved. Otherwise, how do the authors determine that the sensory cells are responsible for the behavioral phenotype?

2) The BRV TRP channels have been implicated in sensing cool temperatures in adult Drosophila and it is unclear how these genes relate to the Irs reported here. From the 2015 PNAS paper, it appears that BRV1-Gal4 is expressed in the cool-sensing neurons as well as additional sensory neurons in larvae. Separating out the role of these channel families in cool sensation would significantly advance the field and help resolve the different studies. Does a RNAi knock-down of BRV1,2, and 3 in Ir21a cells cause a behavioral or cellular phenotype? How do the authors resolve the loss of cool sensing in brv1 mutant larvae with the lack of a cellular phenotype? Do other brv1 neurons respond to cool? Do brv1 neurons and IR21a neurons represent different cool cell populations or do they overlap in expression and function?

Minimally the authors should address this issue forthrightly in Discussion.

eLife. 2016 Apr 29;5:e13254. doi: 10.7554/eLife.13254.015

Author response

Essential revisions:

There are some concerns however that must be addressed before the manuscript can be published in eLife. While it is clear that IR21a and IR25a are necessary for cool sensing in DOCCs, it is not completely convincing that they themselves are cool receptors. As one expert reviewer points out "In contrast to earlier work showing that ectopic expression of Drosophila TRPA1 or GR28B(D) confers heat sensitivity to various cell types (e.g. Ni et al. Nature 2013 from the same group), this does not seem to be the case for IR21a and IR25a, except for a small response in the already thermosensitive Hot Cell neurons. Therefore, various other plausible scenarios can be envisaged where deletion of these proteins would abolish a cold response, without them being involved in the actual cold sensing. For instance, they could act downstream of the actual cold sensor (e.g. similar to a voltage-gated calcium channel at a sensory nerve ending), or their genetic elimination could merely influence the expression of the actual cold sensor. Similarly, it is suggested that these proteins act in a molecular complex, possibly as heteromultimers, but there is no evidence presented that these IRs interact with each other at the molecular level. Given that IR25a has already recently been implicated in thermosensing (Chen et al. 2015), I feel that, for this paper to represent a sufficient advance to merit publication in eLife, additional evidence supporting the hypothesis of IR21a/IR25a complexes as cold sensor should be provided. Minimally the authors should do an experiment where IR21a is expressed ectopically in thermosensitive neurons lacking a hot response (ectopic expression in a Gr28 mutant) to show robust confirmation of cool sensing. Moreover, that authors should discuss their results in light of these caveats.

As requested, we now demonstrate that Hot-Cell-specific Ir21a expression also confers cool sensitivity in a Gr28b null mutant. These new data are presented in a new Figure 6. The manuscript now states: “Finally, ectopic expression of Ir21a in Gr28b mutant HC neurons, which lack the Gr28b(D) warmth receptor, yields neurons that respond only to cooling (Figure 6).”

In terms of significance, it is important to note that while the Chen et al. paper implicated IR25a in thermosensing, our paper provides a very different view of IR25a’s role in the process. Chen et al. suggest IR25a can play an instructive role in thermosensing, acting as a warmth receptor capable of conferring warmth-sensitivity. Our data suggest that IR25a actually plays a permissive role, not specifically mediating hot, cold or chemical detection, but instead assisting other IRs responsible for mediating specific responses. Consistent with such a permissive role, our data show that IR25a also mediates cool sensing, suggesting that IR25a is not a warmth receptor, per se, but rather a molecule that can facilitate either warm or cold sensing depending on the cellular context. Further consistent with this notion, we show that IR25a does not act alone, but requires another IR to mediate cool sensation, and then show that misexpression of this other IR (IR21a) in the HC neurons can confer cool-sensitivity in an IR25a-dependent manner. Taken together, these data all support the notion that IR25a mediates thermosensing much like it mediates chemosensing, by assisting other IRs that have more cell-specific roles.

To put these data in context, most of what is known about IR25a is based on its role in chemical sensing. IR25a is expressed by hundreds of chemosensory neurons and does not appear to form functional chemoreceptors on its own. Instead, IR25a appears to play a permissive rather than an instructive role, serving as a co-receptor for odor-specific IRs that confer specificity for specific chemicals (Benton et al., 2009; Abuin et al., 2011; Silbering et al., 2011). Thus Chen et al.’s suggestion that Ir25a could confer warmth sensitivity when expressed on its own was unexpected. Chen et al. state that “IR25a misexpression confers temperature-dependent firing of heterologous neurons”, but the neurons they were recording from tonically fire in a warmth-responsive fashion even without IR25a misexpression (Chen et al., Figure 4I). IR25a misexpression increases the temperature co-efficient (Q10) of their spiking from 2 to 4 (Chen et al., Figure 4I). As warming speeds up most processes by similar margins (~90% of biological Q10s are between 1.3 and 5.1), this IR25a-dependent thermosensitivity is rather weak. As we note in our manuscript, the requirement for other IR subunits to mediate warmth detection could explain why misexpression of IR25a alone did not confer strong thermosensitivity.

We have adjusted the discussion of the Chen et al. paper slightly to attempt to emphasize this distinction. The manuscript now states: “Interestingly, IR25a was recently implicated in warmth-responsive resetting of the circadian clock, and suggested to confer warmth-sensitivity on its own, without co-expression of other IRs (Chen et al., 2015). The ability of IR25a to serve as a warmth receptor on its own would be a surprise given both its broad expression and its established role as an IR co-receptor (Abuin et al., 2011). As IR25a misexpression only slightly enhanced the thermosensitivity of an already warmth-responsive neuron (Chen et al., 2015), this raises the alternative possibility that – analogous to cool-sensing – IR25a acts not on its own, but rather as a co-receptor with other IRs involved in warmth-sensing. “

In contrast to earlier work showing that ectopic expression of Drosophila TRPA1 or GR28B(D) confers heat sensitivity to various cell types (e.g. Ni et al. Nature 2013 from the same group), this does not seem to be the case for IR21a and IR25a, except for a small response in the already thermosensitive Hot Cell neurons.

The size of the cool response may appear small because it is superimposed on a very large heat response (and cool inhibition). As shown in Figure 5, when cooled from 20°C to 14°C, wild-type HC neurons show an ~2-fold ∆F/F drop in GCaMP fluorescence, which Ir21a misexpression converts into an ~2 fold increase. While this is not as large as the HC neurons’ heat response, IR21a misexpression is turning a cold-inhibited cell into a cold-responsive cell, with an average ∆F/F of ~100%, similar to or larger than other published GCaMP6m responses. To help emphasize this, we now present the quantification of cool response data in a more traditional bar graph format (mean +/- SEM) in the main figure (Figure 5F) and now present the individual data points in Figure 5—figure supplement 1. The impact of Ir21a misexpression is perhaps easier to see in the experiment requested by reviewers in Figure 6; in this figure, IR21a misexpression in a Gr28b mutant HC neuron yields a cool response in the absence of a heat response.

Therefore, various other plausible scenarios can be envisaged where deletion of these proteins would abolish a cold response, without them being involved in the actual cold sensing. For instance, they could act downstream of the actual cold sensor (e.g. similar to a voltage-gated calcium channel at a sensory nerve ending), or their genetic elimination could merely influence the expression of the actual cold sensor. Similarly, it is suggested that these proteins act in a molecular complex, possibly as heteromultimers, but there is no evidence presented that these IRs interact with each other at the molecular level.

We agree. While all work to date indicates that IRs act as sensory receptors and that IR25a functions only within heteromeric complexes, it would be exciting if this were not the case here. We have added a statement noting this: “All studies to date implicate IRs as receptors for sensory stimuli (Rytz et al., 2013), and our mis-expression studies are consistent with a similar role for Ir21a and IR25a in cool sensation. However, we cannot formally exclude the possibility that they could have indirect, and possibly separate, functions in this process, for example, in regulating the expression or function of an unidentified cool receptor.“

There are two other concerns from the expert reviewers that should be addressed.

1) What is the demonstration that the 3 sensory cells that respond to cool are responsible for the behavioral phenotypes? Complete expression patterns of Ir21a-Gal4, Ir25a-Gal4, and R11F02 are necessary to evaluate this. If Ir21a or R11F02 is only in the 3 cool cells, this is easily resolved. Otherwise, how do the authors determine that the sensory cells are responsible for the behavioral phenotype?

We should have been clearer in stating that this was previously addressed in Klein et al., 2015, PNAS. In the revised manuscript, we have altered the sentence in the Introduction that refers to this work. The original manuscript stated: “At the behavioral level, the DOCCs are critical for mediating larval avoidance of temperatures below ~20˚C, with the thermosensitivity of the avoidance behavior paralleling the thermosensitivity of DOCC physiology (Klein et al., 2015).” The revised manuscript now states: “A combination of laser ablation, calcium imaging and cell-specific inhibition studies were used to establish the DOCCs as critical for mediating larval avoidance of temperatures below ~20˚C, with the thermosensitivity of the avoidance behavior paralleling the thermosensitivity of DOCC physiology (Klein et al., 2015).”

To summarize, we showed that laser snipping the axons that connect the Dorsal Organs (DOs) to the brain totally eliminated cool avoidance (Klein et al., 2015; Figure 3B). This showed both that the DO is essential for the behavior and, in addition, that sensory neurons outside the DO are insufficient to support even weak cool avoidance. Calcium imaging was then used to identify 3 cool sensing neurons within each DO (the DOCCs) and a Gal4 expressed in the DOCCs, R11F02-Gal4, was identified (Klein et al., 2015; Figure 3H). We found that inhibiting R11F02-Gal4 neurons eliminated cool avoidance (Klein et al. Figure 3G), and that optogenetically activating them elicited the same behaviors as cooling (Klein et el. Figure 5C-D). In the current manuscript, we extend this prior work, identifying a second DOCC-expressed driver, Ir21a-Gal4 (see Figure 1, which demonstrates that Ir21a-Gal4 is expressed in the DOCCs but no other DO neurons), and find that expression of an Ir21a cDNA under Ir21-Gal4 control rescues cool avoidance. Taken together, the combined use of laser microsurgery and genetics establish the behavioral importance of the DOCCs.

While laser ablation studies indicate that sensory neurons outside the Dorsal Organ are insufficient to support cool avoidance, the requested full animal imaging of both R11F02-Gal4 and Ir21a-Gal4 expression is provided in Figure 1—figure supplement 1. (Ir25a-Gal4 was not used in the paper.). Outside the Dorsal Organ, both Gal4s are expressed by ~100 cells in the brain and ventral ganglion, neurons along the larval body wall and in the tail. R11F02-Gal4 is also expressed by sensory neurons in the Terminal Organ, which laser ablation shows is not required for cool avoidance behavior (Klein et al., 2015; Figure 3B). Again, please note that the role of the Dorsal Organ (and hence the DOCCs) in the behavior does not rely on either R11F02-Gal4 or Ir21a-Gal4 being exclusively expressed in the Dorsal Organ, but was established using a combination of genetics and laser ablation.

2) The BRV TRP channels have been implicated in sensing cool temperatures in adult Drosophila and it is unclear how these genes relate to the Irs reported here. From the 2015 PNAS paper, it appears that BRV1-Gal4 is expressed in the cool-sensing neurons as well as additional sensory neurons in larvae.

NP4486-Gal4 (BRV1-Gal4) is a Gal4 enhancer trap inserted 2kb downstream of the Brv1 gene and 2.5kb upstream of the non-coding RNA gene CR32207 (http://flybase.org/reports/FBti0035982.html). Two recent RNA-seq studies suggest that NP4486-Gal4’s expression pattern appears to reflect CR32207 rather than Brv1 (Menuz et al. 2014 and Shiao et al. 2013). Specifically, while NP4486-Gal4 is widely expressed in the adult antenna (Gallio et al., 2011)), neither RNA-seq study detected any Brv1 transcripts in the antenna. Instead, both detected robust CR32207 expression. These data suggest that NP4486-Gal4 reports on CR32207. This is reasonable, as NP4486-Gal4 is inserted upstream of CR32207’s promoter, a common location for enhancer traps.

Separating out the role of these channel families in cool sensation would significantly advance the field and help resolve the different studies. Does a RNAi knock-down of BRV1,2, and 3 in Ir21a cells cause a behavioral or cellular phenotype? How do the authors resolve the loss of cool sensing in brv1 mutant larvae with the lack of a cellular phenotype? Do other brv1 neurons respond to cool? Do brv1 neurons and IR21a neurons represent different cool cell populations or do they overlap in expression and function?

Minimally the authors should address this issue forthrightly in Discussion.

We tested the roles of brv1 and brv2 in cool sensing by using previously published genetic null mutants rather than RNAi, as RNAi can have off-target effects and also tends to knock down rather than knock out the gene of interest. There are no Brv3 mutants. As shown in Figure 2—figure supplement 2, Brv1 null mutants have a cool avoidance defect, but their DOCC neurons still respond to cold normally. A mutation can cause a behavioral defect in many ways, but whatever Brv1’s role in cool avoidance, Brv1 is not required for DOCC cool sensing. Brv2 mutants show normal cool avoidance. Together our data indicate that IR-mediated cool sensing is independent of these receptors. The manuscript states: “Thus we detect no role for these receptors in cool sensing by the DOCCs.”

We have also added the following sentence to the Discussion:“Interestingly, IR21a- and IR25a-dependent cool sensation appears independent of Brivido 1 and Brivido 2, two TRP channels implicated in cool sensing in the adult (Gallio et al., 2011).”

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

RESOURCES