Expression of the Gr28 genes in the larval sensory organs

A) Schematic representation of larval chemosensory system (Kwon et al., 2011). Three external sensory organs (Dorsal Organ, DO; Terminal Organ, TO; Ventral Organ, VO) hold collectively the dendritic projections of neuronal cell bodies in the respective ganglia (DOG, TOG and VOG). Three clusters of sensory neurons (Dorsa/Ventral Pharyngeal Sensory organ, D/VPS; and Posterior Pharyngeal Sensory organ, PPS) align the pharynx. The antennal nerve (AN) connects the DO neurons to the subesophageal zone (SEZ). The TO and VO neurons project along the maxillary nerve (MN) to the SEZ. The DPS and PPS neurons innervate to the SEZ via labral nerve (LRN). The VPS neurons project to the SEZ through the labial nerve (LBN). Note that olfactory neurons are omitted in the schematic here and in B (below).

B) Expression of the Gr28 genes in the external sensory organs. Note that images show only one of the bilaterally symmetrical organs. Possible co-expression between Gr66a and different Gr28b genes (in B and C; left panels) was assessed using GAL4 and LexA drivers for Gr28 genes and Gr66a, respectively. Expression summary (diagram on the right) shows only one side of the bilaterally symmetrical organs. The Gr28b genes and Gr28a are expressed in different neurons. Likewise, no co-expression is observed between Gr66a (marker for bitter GRNs) and Gr28a, whereas the Gr28b genes are co-expressed in a subset of Gr66a neurons.

C) Expression of the Gr28 genes in internal taste organs. Note that images for the DPS/VPS show both side of the bilaterally symmetrical ganglion, while images of the PPS show only one side of the bilaterally symmetrical organ. Gr28a and Gr66a are partially co-expressed, but each gene is also expressed exclusively in a subset of neurons in the PPS. Gr28bc but none of the other Gr28b genes is co-expressed with Gr66a in a single neuron in the DPS/VPS. Expression summary (diagram on the right) shows only one of the bilaterally symmetrical organs. Immunostaining with anti-GFP (green) and anti-mCD8 (magenta) antibodies was carried out on whole-mount preparations from larvae heads of the following genotypes: (1) UAS-mCD8:RFP lexAop-rCD2:GFP;Gr66a-LexA/Gr28a-GAL4 or Gr28bc-GAL4 or Gr28be-GAL4 (2) UAS-mCD8:RFP lexAop-rCD2:GFP;Gr66a-LexA/+; Gr28ba-GAL4/+. Asterisks refer to a GRN expressing both Gr66a and the indicated Gr28 gene. Scale bars are 5 μm.

Taste behavior of larvae mediated by different Gr28 GRNs

A) Gr28a GRNs mediate capsaicin preference, while Gr28bc GRNs elicit capsaicin avoidance in larvae. Control larvae (w1118, UAS-VR1/+ and Gr (indicated Gr genes)-GAL4/+) show no preference for or avoidance to 0.1 mM (top) or 0.5 mM (bottom) capsaicin. When expressing UAS-VR1 in Gr28a GRNs (Gr28a-GAL4/UAS-VR1), larvae display robust taste preference for capsaicin for both 0.1 and 0.5 mM capsaicin. In contrast, when expressed in Gr28bc GRNs (Gr28bc-GAL4/UAS-VR1), larvae strongly avoid both 0.1 and 0.5 mM capsaicin. Neither avoidance nor preference was observed when VR1 was expressed in the Gr28ba or Gr28be GRNs. For comparison, expression of VR1 in sweet taste neurons (Gr43aGAL4/UAS-VR1) or bitter taste neurons (Gr66a-GAL4/UAS-VR1) led to strong preference for or strong avoidance of capsaicin. Each bar represents the mean ± SEM of P.I. (n = 11-42 assays). The taste behavior of Gr-GAL4 > UAS-VR1E600K larvae is compared with 3 controls (w1118, UAS-VR1E600K/+ & Gr-GAL4/+) by using one-way ANOVA with Bonferroni’s multiple comparison tests (p < 0.05), whereby different letters indicate statistically significant difference. Dashed lines delineate groups for ANOVA. A short solid line indicates control groups.

B) Gr28aonly GRNs are necessary and sufficient for inducing capsaicin preference of larvae. When VR1 expression is restricted to GRNs not co-expressing Gr66a or Gr28bc (Gr28aonly GRNs) by means of lexAop-GAL80 under control of Gr66a-LexA, larvae show strong preference for capsaicin. Each bar represents the mean ± SEM of P.I. (n = 13-18 assays). Bars with different letters are significantly different (one-way ANOVA with Bonferroni’s multiple comparison tests (p < 0.05). Fly genotypes: wild-type: w1118 (black), w1118;UAS-VR1E600k/+ (light gray), w1118;Gr28a-GAL4/+ (dark grey), w1118;Gr28a-GAL4/UAS-VR1E600K (white), w1118;+;Gr28ba-GAL4/+ (dark grey), w1118;UAS-VR1E600K/+;Gr28ba-GAL4/+ (white), w1118;Gr28bc-GAL4/+ (dark grey), w1118;Gr28bc-GAL4/UAS-VR1E600K (white), w1118;Gr28be-GAL4/+ (dark grey), w1118;Gr28be-GAL4/UAS-VR1E600K (white), w1118;Gr43aGAL4/+ (dark grey), w1118;Gr43aGAL4/UAS-VR1E600K (white), w1118;Gr66a-GAL4/+ (dark grey), w1118;Gr66a-GAL4/UAS-VR1E600K (white), w1118;UAS-VR1E600k/+;lexAop-GAL80/+ (light gray), w1118;Gr66a-LexA/+;Gr28a-GAL4/+ (dark grey) and w1118;Gr66a-LexA/UAS-VR1E600K;Gr28a-GAL4/lexAop-GAL80 (white).

The Gr28b neurons mediate avoidance behavior to bitter compounds

Inactivation of Gr28bc GRNs (Gr28bc-GAL4/UAS-Kir2.1) elicits significantly reduced avoidance of larvae to all four bitter compounds tested: denatonium, quinine, lobeline and caffeine, while control larvae carrying either the driver or the reporter only showed strong avoidance of these compounds. In contrast, flies with inactivated Gr28be GRNs (Gr28be-GAL4/UAS-Kir2.1) still avoid denatonium and quinine (top), but no longer avoid lobeline and caffeine (bottom). Each bar represents the mean ± SEM of P.I. (n = 11-20 assays). The taste behavior of Gr-GAL4 > UAS-Kir2.1 larvae is compared with 2 controls (UAS-Kir2.1/+ & Gr28b-GAL4/+) by using Kruskal-Wallis test by ranks with Dunn’s multiple comparison tests (p < 0.05). Bars with different letters are significantly different. Dashed lines delineate groups for ANOVA. A short solid line indicates a control. Fly genotypes: w1118; UAS-Kir2.1/+ (light grey), w1118; Gr28bc-GAL4/+ (dark grey), w1118; Gr28bc-GAL4/UAS-Kir2.1 (white), w1118; Gr28be-GAL4/+ (dark grey), w1118; Gr28be-GAL4/UAS-Kir2.1 (white).

Role of individual Gr28 genes in bitter taste avoidance

A) Some of the Gr28 genes are strictly required for sensing denatonium. Wild-type (w1118) larvae, but not Gr28 mutant larvae, strongly avoid denatonium. In contrast, Gr28 mutant (w1118;ΔGr28/ΔGr28) larvae do not show significantly reduced avoidance of quinine, lobeline and caffeine. Each bar represents the mean ± SEM of P.I. (n = 12-22 assays). Asterisks indicate a significant difference between the Gr28 mutant and wild-type larvae (Two-tailed, Mann-Whitney U test, **** p < 0.0001, *** p < 0.001, ns: not significant).

B) Single Gr28b genes can rescue avoidance response to denatonium when expressed in Gr28bc neurons of Gr28 mutant larvae. Each bar represents the mean ± SEM of P.I. (n = 11-22 assay). The behavior of Gr28 mutant larvae expressing UAS-Gr28 transgenes under control of the Gr28bc-GAL4 driver was compared to Gr28+/+control (w1118, filled bar), and three Gr28 mutant controls (ΔGr28, ΔGr28 plus driver, and ΔGr28 plus respective UAS-Gr28 reporter, plain bar) using Kruskal-Wallis test by ranks with Dunn’s multiple comparison tests (p < 0.05). Bars with different letters are significantly different. Fly genotypes: wild-type: w1118(black), mutants: w1118;ΔGr28/ΔGr28, w1118;ΔGr28/ΔGr28 Gr28bc-GAL4, w1118;ΔGr28/ΔGr28;UAS-Gr28 (indicated Gr28 genes)/+ and w1118;ΔGr28/ΔGr28 UAS-GCaMP6m;UAS-Gr28 (for Gr28bb or Gr28bc genes)/+ (white), rescues: w1118;ΔGr28/ΔGr28 Gr28bc-GAL4;UAS-Gr28 (indicated Gr28 genes)/+ and w1118;ΔGr28 UAS-GCaMP6m/ΔGr28 Gr28bc-GAL4;UAS-Gr28 (for Gr28bb or Gr28bc genes)/+ (grey).

Cellular activation of Gr28bc GRNs by bitter compounds that require Gr28ba or Gr28bc function.

A)Diagram of Ca2+ imaging experimental set up.

B) Representative still images of Ca2+ response in the Gr28bc expressing TO neuron. Ca2+ responses of the Gr28bc GRNs upon stimulation with indicated ligands. ΔF indicates the changes in fluorescence light intensity of the cell body before/after ligand application.

C & D) Representative traces (C) and quantified Ca2+ responses (D) of the Gr28bc GRNs after stimulation with indicated ligands. Fly genotype: w1118; Gr28bc-GAL4/UAS-GCaMP6m. Each bar represents the mean ± SEM of Ca2+ imaging with 12 – 16 larvae. Asterisks indicate a significant difference between carrier (water) and indicated ligands (Two-tailed, Mann-Whitney U test, *** p < 0.001, ns: not significant).

(E) Neurons of larvae lacking the Gr28 genes exhibit significantly reduced responses to denatonium and quinine. The Gr28bc expressing TO GRN of Gr28 mutant larvae (ΔGr28) display significantly reduced Ca2+ responses to denatonium and quinine but not to lobeline or caffeine when compared to Gr28bc expressing TO of wild type controls. Responses to denatonium, but not quinine, is restored by expression of single Gr28b (Gr28ba or Gr28bc) genes (grey). Larvae genotypes: Gr28+ control (black bar): w1118; Gr28bc-GAL4/UAS-GCaMP6m. Gr28- control (white bar): w1118; ΔGr28 Gr28bc-GAL4/ΔGr28 UAS-GCaMP6m. Each bar represents the mean ± SEM with 13 – 16 larvae. Asterisks indicate a significant difference between Gr28+control and Gr28 mutant (ΔGr28) larvae (Two-tailed, Mann-Whitney U test, *** p < 0.001, ** p < 0.01 ns: not significant).

(F and G) Gr28bc or Gr28ba rescues denatonium responses in GRNs of ΔGr28 larvae. Expression of Gr28bc or Gr28ba in under control of Gr28bc-GAL4 restores responses to denatonium, but not quinine in GRNs of ΔGr28 larvae. Each bar represents the mean ± SEM of Ca2+ imaging with 12 – 17 larvae. The Ca2+ responses of Gr28 mutant larvae expressing UAS-Gr28 transgenes under Gr28bc-GAL4 driver is compared to Gr28+ (black) and Gr28-(white) controls by using Kruskal-Wallis test by ranks with Dunn’s multiple comparison tests (p < 0.05). Bars with different letters are significantly different. Dashed lines delineate groups for ANOVA. A short solid line indicates control groups. Fly genotypes: Gr28+ control (black bar): w1118; Gr28bc-GAL4/UAS-GCaMP6m. Gr28- control (white bar): w1118; ΔGr28 Gr28bc-GAL4/ΔGr28 UAS-GCaMP6m. Gr28 rescues (grey bar): w1118; ΔGr28 Gr28bc-GAL4/ΔGr28 UAS-GCaMP6m; UAS-Gr28/+. Concentrations of ligands were 100 mM for sugars, 50 mM for caffeine and 5 mM for denatonium, quinine and lobeline.

Role of different GRN subsets in taste behavior of larvae

GRNs mediating avoidance behavior express Gr28bc, while GRNs necessary for feeding preference express Gr28a alone (Gr28aonly GRNs). Note that each ensemble is composed of at least a pair of neurons located in the external taste organs and a pair of neurons in the internal taste organs.

Amino acid alignment of the six Gr28 proteins

The sequence alignment was generated using Clustal Omega tool from ClustalW2 (https://www.ebi.ac.uk/Tools/msa/clustalo/). IL and EL indicate intracellular loop and extracellular loop, respectively. Note that the C terminal region starting at the IL3 is identical in the Gr28b proteins. Red highlighted letters indicate amino acids identical only in Gr28ba and Gr28bce. Green highlighted letters indicate amino acids conserved in Gr28ba, Gr28bc and one other Gr28 protein. An asterisk below the sequences indicates residues identical in all Gr28 proteins; a colon a strongly conserved residue (STA, NEQK, NHQK, NDEQ, QHRK, MILV, MILF, HY, FYW); and a period moderately conserved residue (CSA, ATV, SAG, STNK, STPA, SGND, SNDEQK, NDEQHK, NEQHRK, FVLIM, HFY). TM1-7 indicates helical transmembrane segment predicted by HMMTOP 2.0 software (http://www.enzim.hu/hmmtop).

Identification of candidate amino acids for ligand binding pockets of Gr28bc.

Top: Sixteen amino acids conserved in both Gr28ba and Gr28bc are mapped on 3D structure of Gr28bc using Alphafold (https://www.alphafold.ebi.ac.uk/). Red circles indicate amino acid locations of residues identical in Gr28ba and Gr28bc. Green circles indicate amino acid locations of residues identical in one additional Gr28 protein (Gr28bb or Gr28a). Bottom: Prediction of putative ligand binding pocket in of Gr28bc with highest probability using Prankweb (https://prankweb.cz/). The three resides part of this pocket and identical between Gr28ba and Gr28bc are indicated in green (V177, I218 & I222). The color displayed in the protein structure indicates a per-residue confidence score (pLDDT) between 0 and 100 as predicted by Alphafold.

Conserved amino acids in the amino termini of Gr28ba and Gr28bc

Prediction of ligand binding sites from protein structure of Gr28bc

Co-expression analysis between Gr28 genes in larval sensory organs

Immunostaining with anti-GFP (green) and anti-mCD8 (magenta) antibodies on whole-mount preparations of the heads from larvae of the genotypes: (1) UAS-mCD8:RFP lexAop-rCD2:GFP;Gr28a-GAL4 or Gr28be-GAL4 or Gr66a-GAL4/+;Gr28bc-LexA/+: (2) UAS-mCD8:RFP lexAop-rCD2:GFP;+; Gr28ba-GAL4/Gr28bc-LexA. Asterisks refer to GRN expressing both Gr28bc and the indicated Gr66a or Gr28 gene. Note that images show only one of the bilaterally symmetrical external organs in the DOG/TOG (A) and both bilaterally symmetrical organs in the DPS/VPS (B).

Gr66a gene is necessary for caffeine avoidance of larvae

Gr66a null mutant larvae are unable to avoid of caffeine but none of other bitter compounds. Each bar represents the mean ± SEM of P.I. (n = 11-12 assays). Asterisks indicate a significant difference between the Gr66a mutant (Gr66aex83) and control larvae (w1118) (Two-tailed, Mann-Whitney U test, ** p < 0.01, ns: not significant).