The Basis of Food Texture Sensation in Drosophila

Abstract

Food texture has enormous effects on food preferences. However, the mechanosensory cells and key molecules responsible for sensing the physical properties of food are unknown. Here, we show that akin to mammals, the fruit fly, Drosophila melanogaster, prefers food with a specific hardness or viscosity. This food texture discrimination depends upon a previously unknown multidendritic (md-L) neuron, which extends elaborate dendritic arbors innervating the bases of taste hairs. The md-L neurons exhibit directional selectivity in response to mechanical stimuli. Moreover, these neurons orchestrate different feeding behaviors depending on the magnitude of the stimulus. We demonstrate that the single Drosophila transmembrane channel-like (TMC) protein is expressed in md-L neurons, where it is required for sensing two key textural features of food-hardness and viscosity. We propose that md-L neurons are long sought after mechanoreceptor cells through which food mechanics are perceived and encoded by a taste organ, and that this sensation depends on TMC. VIDEO ABSTRACT.

Figure 1. Feeding responses to liquid and solid foods with different viscosities and hardness, respectively

(A) Viscosities (centipoise) of solutions containing different concentrations of HPC (0—1.5%). (B) Tip recordings of wild-type L4 sensilla showing the electrophysiological responses to either 100 mM sucrose or 1% hydroxypropyl cellulose (HPC). (C) Average number of spikes produced in wild-type L4 sensilla in response to 0—1.5% HPC plus 100 mM sucrose or 0—1.5% HPC alone. n=5. (D—F) Fly immobilized in a pipet tip for the proboscis extension reflex (PER) assay. (D) Prior to presentation of a food stimulus to the fly. (E) Fly stimulated with a drop of a 100 mM sucrose solution. (F) Fly stimulated with a drop of a 1.5% HPC solution mixed with 100 mM sucrose. (G) PER of control animals in response to stimulation with 0—1.5% HPC mixed with 100 mM sucrose. n=20. (H) Rheological measurements of different percentages of agarose mixed with 2 mM sucrose. Stiffness was calculated by dividing the mechanical stress by the strain of the agarose gels. The stiffness unit is newton/meter² (pascal; Pa). (I) Gustatory preferences exhibited by wild-type flies as a function of the agarose concentration. To perform the two-way feeding assays, one side of the Petri dish was filled with food made up of 1% agarose gel and 2 mM sucrose, and the other side contained 2 mM sucrose and agarose gels of varying percentages (0.5—4%) as indicated. ~70 adult flies/assay. n=5 trials. (J) Percentages of the PER exhibited by control flies in response to stimulation with 0.5—4% agarose-containing food mixed with 100 mM sucrose. n=20. The error bars indicate SEMs. *p<0.05. ANOVA tests with Scheffé’s post-hoc analysis.

Figure 2. tmc mutant flies displayed impaired feeding responses to foods with different viscosity or hardness

(A) Screening of candidate receptors and channels for impacts on food selection based on hardness. The indicated animals were allowed to choose between 2 mM sucrose mixed with either 1% or 3% agarose using two-way food preference assays. The controls were w¹¹¹⁸ flies. n=5 trials. ~70 flies for each trial. (B) Genomic structures of the tmc gene and the strategy for making tmc¹ knock-out flies by homologous recombination. The DNA fragment (red box) encoding the TMC domain was replaced by the mini-white⁺ (w⁺) gene. To generate the tmc promoter Gal4 line, we used the indicated 3 kb genomic DNA fragment flanking the 5’ end of the transcriptional start site of tmc. (C) Two-way feeding preferences exhibited by flies offered 1% agarose + 2 mM sucrose versus 2 mM sucrose containing various concentrations of agarose. The “rescue” flies in C—E are tmc-Gal4/UAS-tmc;tmc¹. n≥5 trials. (D) The percentages of PER in response to stimulation with food containing 100 mM plus 0.5—4% agarose. n=20. (E) The percentages of PER in response to presentation of 100 mM sucrose and 0—1.5% HPC. n=20. The error bars indicate SEMs. *p<0.05. ANOVA tests with Scheffé’s post-hoc analysis.

Figure 3. Expression pattern of tmc in the peripheral taste organ and the central brain

The staining patterns were viewed by confocal microscopy. (A and B) Staining of labella using TMC antibodies. (A) Control (w¹¹¹⁸). (B) tmc¹. (C) Anti-GFP staining of a labellum from flies expressing the UAS-mCD8::GFP reporter under control of the tmc-Gal4. (D) The relative positions of md-L dendritic arbors (visualized by anti-GFP) and taste sensilla decorating the labella. Shown is a confocal image of anti-GPF staining superimposed on an image obtained by differential interference contrast. Representative S-, I- and L-type sensilla are marked. The I6 sensillum is indicated. (E) Anti-GPF staining showing the expression pattern of the tmc-Gal4 in the internal mouth. (F) Double-labeling of Gr5a GRNs and md-L neurons in a labellum from tmc-QF/QUAS-mCD8::GFP;Gr5a-Gal4/UAS-DsRed flies. Green, anti-GFP; red, DsRed. (G) A close-up view of the relative locations of md-L dendritic terminals (green), Gr5a GRN soma and dendrites (red) from tmc-QF/QUAS-mCD8::GFP;Gr5a-Gal4/UAS-DsRed flies. (H) Central brain expression pattern of the tmc-Gal4/UAS-mCD8::GFP reporter. The axons of md-L neurons (green) project from the proboscis to the SEZ. Anti-nc82 (red) stains the active zones of neurons throughout the brain. AL, antennal lobe; MB, mushroom body; SEZ, subesophageal zone. (I) Relative projection patterns of axons extending from Gr5a GRNs (green) and md-L neurons (red) into the SEZ of tmc-QF/QUAS-tdTomato;Gr5a-Gal4/UAS-mCD8::GFP flies. The scale bars indicate 20 μm in all panels except for 10 μm in G and 40 μm in panel H.

Figure 4. Electrophysiological responses of md-L neurons to mechanical stimuli

(A) Bending of L-type taste bristles exerted compression forces underneath dendritic arbors of md-L neurons, thereby leading to the firing of action potentials in md-L neurons. L4 taste bristles from the indicated flies were deflected 20 μm in the dorsal, ventral, anterior and posterior directions. The rescue flies were tmc-Gal4/UAS-tmc;tmc¹. The vertical arrows indicate the onset of the mechanical stimuli. (B) Box and whisker plots showing the number of spikes/500 ms. n=10. (C—E) Tip recording traces showing the responses of control and tmc¹ GRNs to chemical stimulation. The box and whisker plots show the summary tip recording data. Not significant, n.s. n=5. (C) 50 mM NaCl (L4 sensilla). (D) 50 mM sucrose (L4 sensilla). (E) 10 mM caffeine (S6 sensilla). The error bars indicate SEMs. *p< 0.05. ANOVA tests with Scheffé’s post-hoc analysis.

Figure 5. Ca²⁺ dynamics of tmc neurons in response to mechanical stimuli

(A and B) Representative images showing the relative changes in Ca²⁺ levels in the SEZ with and without mechanical stimulation of the labellum. The Ca²⁺ dynamics were monitored using GCaMP6f. The color scale to the right of (H) shows ΔF/F. (A) tmc-Gal4,UAS-GCaMP6f. (B) tmc-Gal4,UAS-GCaMP6f;tmc¹. (C) The time course showing the Ca²⁺ dynamics in the SEZ of tmc-Gal4,UAS-GCaMP6f and tmc-Gal4,UAS-GCaMP6f;tmc¹ flies. Shown are the fold changes in fluorescence intensity (ΔF/F₀). The arrow indicates the onset of mechanical stimuli. n=5. (D) The fold changes in peak fluorescence intensity (ΔF/F₀) in response to different deflection distances for tmc-Gal4,UAS-GCaMP6f and tmc-Gal4,UAS-GCaMP6f;tmc¹ flies. n=5. The error bars indicate SEMs. *p<0.01. ANOVA tests with Scheffé’s post-hoc analysis.

Figure 6. Requirements for md-L neurons for sensing food hardness and viscosity

(A and B) Intersectional genetic labeling of the central projections of md-L neurons. (A) Central brain expression pattern of tmc-Gal4,UAS-mCD8::GFP. (B) Central brain expression pattern of tmc-Gal4,UAS-mCD8::GFP;vGluT-Gal80. (C) PER responses to foods containing 100 mM sucrose plus either 3% agarose or 1.5% HPC. Shown are the responses of controls and flies expressing the indicated transgenes. n=20. (D and E) Images showing the soma and axon of an md-L neuron (tmc-Gal4,UAS-DsRed). (D) Prior to laser treatment. (E) After laser treatment. (F) PER responses to foods containing 100 mM sucrose and either 3% agarose or 1.5% HPC before and after md-L neurons were exposed to laser treatments. n=15. The error bars indicate SEMs. *p<0.05. ANOVA tests with Scheffé’s post-hoc analysis.

Figure 7. Optogenetic activation of md-L neurons is sufficient to trigger proboscis responses

(A–C) Dynamics of light-induced proboscis extension in a fly expressing tmc-Gal4,UAS-CsChrimson. (A) Prior to light stimulus. (B) Full proboscis extension triggered by a moderate level of light (0.4 mW/mm²) (C) Post-light stimuli. (D) Lack of proboscis extension response upon exposure of control flies to weak light (0.1 mW/mm²). (E) Lack of proboscis extension response upon exposure of control flies to strong light (1.0 mW/mm²). (F) tmc-Gal4/UAS-CsChrimson flies extend their proboscis in response to weak light (0.1 mW/mm²). (G) tmc-Gal4/UAS-CsChrimson flies retract their proboscis in response to strong light (1.0 mW/mm²). (H) Relationship between the intensity of the stimulating light and percentages of the PER using the indicated flies. n≥15. (I) Percentages of flies of the indicated genotypes showing PERs in response to weak (0.1 mW/mm²) or strong light stimuli (1 mW/mm²). n≥15. The error bars indicate SEMs. *p<0.01. ANOVA tests with Scheffé’s post-hoc analysis.

Figure 8. Differential modulation of sucrose feeding with varying intensities of light

(A and B) PERs displayed by a control fly presented with 500 mM sucrose solution under: (A) weak light stimulation (0.1 mW/mm²), and (B) strong light stimulation (1.0 mW/mm²). (C) PER exhibited by a fly expressing tmc-Gal4,UAS-CsChrimson presented with 500 mM sucrose solution under weak light stimulation (0.1 mW/mm²). (D) No or minimal PER elicited by a fly expressing tmc-Gal4,UAS-CsChrimson presented with 500 mM sucrose solution under strong light stimulation (1.0 mW/mm²). (E) Relationship between the intensity of the stimulating light and PER percentages using the indicated flies. n≥15. (F) Percentages of flies of the indicated genotypes showing PERs in response to weak (0.1 mW/mm²) or strong light stimuli (1 mW/mm²). n≥15. The error bars indicate SEMs. *p<0.05. ANOVA tests with Scheffé’s post-hoc analysis.