Neuroscience

Optogenetic induction of appetitive and aversive taste memories in Drosophila

Meghan Jelen
Pierre-Yves Musso
Pierre Junca
Michael D. Gordon author has email address

Department of Zoology and Life Sciences Institute, University of British Columbia, Vancouver, Canada

https://doi.org/10.7554/eLife.81535.1

Open access
Copyright information

Figures and data

GRNs produce punishment and reward signals capable of facilitating taste memory formation.
(A) Diagram outlining STROBE memory paradigm. Training: starved flies freely interact with a LED-activating tastant for 40 minutes. CsChrimson induces bitter or sweet neuron stimulation upon LED-activation, pairing feeding with punishment or reward. Testing: associative memory is measured by assessing preference for the tastant compared to agar for a 1-hour time period. (B) Schematic of aversive and appetitive STM and LLM timelines and a diagram of activated bitter and sweet neurons in the SEZ. (C) Preference indices (left) and tastant interaction numbers (right) for Gr66a>CsChrimson flies compared to genetic controls during training and 10 minutes later upon testing. (D) Cumulative average preference indices over the course of testing in (C), (n=16-30). (E) Preference index (left) and interactions numbers (right) for Gr43a>CsChrimson flies fed all-trans-retinal compared to controls in the short-term memory assay. (F) Preference index of flies in (E) throughout testing (n= 12-23). (G) Preference index (left) and interactions numbers (right) for Gr43a>CsChrimson flies fed all-trans-retinal compared to controls in the long-lasting memory assay. (H) Average preference index as a function of time for the testing period of flies in the long-lasting memory assay (n= 14-30). Preference index is mean ± SEM, One-way ANOVA, Dunnett’s post hoc test: **p < 0.01, ****p < 0.0001.

PPL1 and PAM neural activation is sufficient for the induction of short and long-lasting taste memories.
(A, B) Paradigm timeline schematic and MB model indicating PPL1 compartments activated by optogenetics. Preference indices for MB504B>CsChrimson flies and controls as an average (A), and over time (B) in the short-term memory assay (n=19-31). (C, D) Schematic for the long-lasting memory assay, and preference index for MB504B>CsChrimson flies compared to genetic controls as an average (C), and corresponding time curve (D) for the duration of testing (n=20-33). (E, F) Assay timeline, and MB model with activated PAM compartments highlighted. Preference index for R58E02>CsChrimson flies comparted to controls (E) and as a function of time (F) in the short-term memory assay (n=25-38). (G, H) Preference index for R58E02>CsChrimson flies compared to controls (E) and over the total duration of testing (F) in the long-lasting memory assay (n=17-35). Preference indices are mean ± SEM, One-way ANOVA, Dunnett’s post hoc test: *p < 0.05, **p < 0.01, ***p < 0.001.

The MB is required for the formation of short- and long-term taste memories
(A, B) Preference indices for Gr43a>CsChrimson flies in the short-term (n=16-34) (A) and long-lasting (n=13-27) (B) memory assays when the MB is silenced, compared to controls. (C, D) Preference indices for R58E02>CsChrimson flies when the MB is silenced in the short-term (n=24-28)(C) and long-lasting (n=17-23) (D) memory paradigms, compared to controls. (E) Preference indices in the long-lasting memory assay for R58E02>CsChrimson flies fed protein synthesis inhibitor cycloheximide compared to vehicle-fed controls (n=17-22). (F) Model of appetitive taste memory formation via GRN/PAM activation. Preference indices are mean ± SEM, t-test/One-way ANOVA, Dunnett’s post hoc test: *p < 0.05, **p < 0.01.

Discrete non-overlapping PAM subpopulations induce appetitive short- and long-term taste memories.
(A, B) PAM subpopulation R48B04 innervates highlighted MB compartments (left), and preference indices of R48B04>CsChrimson flies for 75mM NaCl is tested in the short-term (n=21-28) (A), and long-term (n=15-17) (B) memory assays with or without retinal. (C, D). PAM subpopulation R15A04 innervates non-overlapping MB sub-regions compared to R48B04. Preference indices for R15A04>CsChrimson flies fed all-trans-retinal in the short-term (n=11-15) (C) and long-term (n=20-27) (D) taste memory assays with or without retinal. (E, F) PAM-α1 innervates a single compartment in the MB. Preference indices of MB043B>CsChrimson flies in the short-term (n=11-14) (E) and long-term (n=19-22) (F) memory assays with or without retinal. (G, H) PAM-β2β′2a synapses on the highlighted MB compartment. Preference indices for MB301B>CsChrimson flies during the short-term (n=20-27) (G) and long-term (n=10-15) (H) memory assays with or without retinal. Preference indices are mean ± SEM, t-test: **p < 0.01.

Caloric food sources are required for the formation of associative long-term taste memories.
(A) Graphic of the timeline followed for the long-term taste memory paradigm, and the MB compartments innervated by PAM driver R58E02-Gal4. (B, C) Preference indices for R58E02>CsChrimson and control flies during training (B), and testing (C), after being refed with a caloric or non-caloric medium (n=13-28). Preference indices are mean ± SEM, One-way ANOVA, Dunnett’s post hoc test: *p < 0.05, **p < 0.01.

MB-MP1 neuron activation post-training replaces energy signal required for the formation of LTM.
(A) Graphic of timeline followed for the LTM taste assay with thermogenetic activation of MB-MP1 neurons. Preference indices during training and testing for R58E02>CsChrimson flies fed all-trans-retinal, with MB-MP1 neurons thermogenetically activated post training using R30E11>TRPA1, compared to controls without MB-MP1 activation (n=18-29). (C) Schematic depicting summary of research. Preference indices are mean ± SEM, One-way ANOVA, Dunnett’s post hoc test: **p < 0.01.

Gustatory receptor neuron activation produces reward and punishment signals in taste memory formation.
(A) Projections from GR66a, Gr43a, PPK23^Glut and GR64f GRNs. (B) Preference index for Gr66a>CsChrimson flies fed all-trans-retinal in the long-lasting taste memory assay with or without retinal (n=19-21). (C) Preference indices for PPK23^Glut>CsChrimson flies in the short-term taste memory assay with or without retinal (n=22-30). (D, E) Preference index for Gr64f>CsChrimson flies in the short-term (n=28-36) (D) and long-lasting (n=13-21) (E) taste memory assay with or without retinal. Preference indices are mean ± SEM, t-test: *p < 0.05, **p < 0.01, ***p < 0.001.

Taste memories are specific to the CS⁺.
(A) Schematic outlining STM assay and MB compartments innervated by PAM driver R58E02. Preference index of R58E02>CsChrimson flies in the STM assay when monopotassium glutamate (MPG) is used as a CS⁺ (n=16-19). (B) Preference index of R58E02>CsChrimson flies in the STM assay with salt as a CS⁺ and MPG as a CS^- (n=16-19). (C) Preference indices of R58E02>CsChrimson flies in the short-term memory assay when the CS⁺ (NaCl) is switched for to the novel tastant (N.T.) MPG during testing (n=21-30). Preference indices are mean ± SEM, t-test, : *p < 0.05.

Activation of discrete PAM subpopulations induces distinct types of taste memories.
(A, B) Schematic outlining STM and LTM assays, and an alternate PAM-α1 driver, innervating the α1 compartment of the horizontal MB lobe. Preference indices of MB299B>CsChrimson flies fed all-trans-retinal in the short-term (n=18-23) (A) and long-term (n=22-28) (B) memory assays compared to controls. Preference indices are mean ± SEM, t-test: **p < 0.01. (n=18-26)

Sign up for email alerts