Tyramine action on motoneuron excitability and adaptable tyramine/octopamine ratios adjust Drosophila locomotion to nutritional state

Abstract

Adrenergic signaling profoundly modulates animal behavior. For example, the invertebrate counterpart of norepinephrine, octopamine, and its biological precursor and functional antagonist, tyramine, adjust motor behavior to different nutritional states. In Drosophila larvae, food deprivation increases locomotor speed via octopamine-mediated structural plasticity of neuromuscular synapses, whereas tyramine reduces locomotor speed, but the underlying cellular and molecular mechanisms remain unknown. We show that tyramine is released into the CNS to reduce motoneuron intrinsic excitability and responses to excitatory cholinergic input, both by tyramine^honoka receptor activation and by downstream decrease of L-type calcium current. This central effect of tyramine on motoneurons is required for the adaptive reduction of locomotor activity after feeding. Similarly, peripheral octopamine action on motoneurons has been reported to be required for increasing locomotion upon starvation. We further show that the level of tyramine-β-hydroxylase (TBH), the enzyme that converts tyramine into octopamine in aminergic neurons, is increased by food deprivation, thus selecting between antagonistic amine actions on motoneurons. Therefore, octopamine and tyramine provide global but distinctly different mechanisms to regulate motoneuron excitability and behavioral plasticity, and their antagonistic actions are balanced within a dynamic range by nutritional effects on TBH.

Keywords: Dmca1D; biogenic amine; calcium channel; insect; neuromodulation.

Fig. 1.

OA/TA-containing neurons adjust locomotor activity and contact MN dendrites. (A) Representative traces of 2 min of crawling filmed at 4 frames/s (*, starting position) from CS larvae after 2 h of starvation (first trace), continuous feeding (second trace), and feeding of TA (third trace) and from a fed TBH mutant (TβH^nM18) animal (fourth trace). Starvation significantly increases (B) crawling distance and (C) speed (dark gray bars). TA significantly reduces locomotor activity (magenta bars). TBH mutants with no OA but increased TA exhibited highly significantly reduced locomotor activity (light gray bars). **P < 0.01; ***P < 0.001; ANOVA with Newman–Keuls post hoc testing. (D–Diii) Maximum projection images of triple labeling of aCC and RP2 MNs with GFP (w;P{eve-GAL4.RN2}P,P{UAS-mcd8-GFP.L}LL5/+; act>>GAL4 UAS-FLP/+) (Di, green), VUM neurons with anti-TDC2 (Dii, magenta), and the presynaptic active zone with anti-NC82 (Brp, Diii, cyan). Dotted white boxes indicate enlargements shown in E–Eiii (total z distance of 5 µm). White arrowheads demark spots with overlap of all three labels. Single optical sections (z = 0.5 µm) from areas in dotted white boxes are enlarged in F–Fiii. Arrows mark VUM neuron central arbor varicosities which are in direct contact with MN dendrites and contain the presynaptic marker Brp.

Fig. 2.

TA reduces MN electrical excitability. (A–Aii) Representative response of a RP2 MN in CS to square-pulse current injection of 80-pA amplitude before (A), 2 min after TA application (10⁻⁵ M) (Ai), and following 2-min washout in saline (Aii). TA significantly and reversibly increases the delay to the first spike (pulse onset to first spike) (B) and decreases firing rate (mean firing rate of response spike train) (C) but has no effects on firing threshold (D) or input resistance (E) (n = 21). Thermogenetic activation of TA-containing neurons (w;tdc2-GAL4,UAS-TRPA1;+ at 30 °C, control at 20°) reversibly decreases RP2 firing rate (F) (n = 9). (G–Gii) Representative responses of RP2 to ramp current injection of 100-pA (black) and 180-pA (gray) amplitude before (G), 2 min after bath application of TA (10⁻⁵ M) (Gi), and following 2 min of washout (Gii). TA significantly and reversibly increases the delay to the fist spike (ramp onset to first spike) (H) and decreases firing rate (mean from first to last spike during ramp; n = 21) (I). These effects remained in the absence of chemical synaptic transmission in shi^ts (w¹¹¹⁸ shi¹) animals at nonpermissive temperature (30 °C) (J). Dose–response tests revealed significant effects of TA at 10⁻⁶ and 10⁻⁵ M but not at 10⁻⁴ M (K). Action potential (AP) shape was not affected by TA (black trace before TA, red trace with 10⁻⁵ M TA) (L). **P < 0.01; ***P < 0.001; Kruskal–Wallis ANOVA with Mann–Whitney U test pairwise comparisons. AHP, after hyperpolarization; n.s., not significant; sal, saline; syn isol, synaptic isolation.

Fig. 3.

TA action on MNs requires the honoka receptor and Dmca1D Ca²⁺ channels. (A–Aii) Ca²⁺ signals in RP2 MN dendrites of a representative control animal (w;OK371-GAL4/20xUAS-IVS-GCaMP6m attP40;+) in response to focal pressure application of nicotine (10⁻⁵ M) through a glass electrode placed into the motor neuropil within ∼10 µm of GCaMP6-labeled MN dendrites. Nicotine was puffed in two trains each before (A), 2 min in TA (gray shaded area) (Ai), and following 2 min washout (Aii). Each train consisted of four to five consecutive puffs (see black arrows) with a 15-s interpuff interval. Intertrain interval was 1 min, and TA wash in and washout durations were 2 min each. Upper row shows original images, and Lower row shows changes in relative GCaMP fluorescence (ΔF/F) over time. (B) TA significantly and reversibly reduces dendritic Ca²⁺ responses. (C and D) TA has no effect on MN dendritic Ca²⁺ responses in honoka mutants (w;OK371-GAL4/20xUAS-IVS-GCaMP6m attP40;oct-tyrR^hono/oct-tyrR^hono). (E and F) TA has no effect on RP2 Ca²⁺ responses to nicotine following RNAi knockdown of Dmca1D in MNs (w;OK371-GAL4/20xUAS-IVS-GCaMP6m attP40;UAS-Dmca1D-RNAi ^HMS00294/+). (G and I) Firing responses of RP2 to somatic ramp current injections in honoka mutants (+;;P{lwB}Oct-TyrR^hono). (G) Recordings before (Upper trace), 2 min in TA (10⁻⁵ M) (Middle trace), and after 2-min washout (Lower trace) indicate no differences. (H) Quantification shows that TA induced reductions in firing frequency in controls but not following bath application of the TA receptor blocker yohimbine or in honoka mutants or following hono RNAi in MNs (w*;OK371-GAL4/+;UAS-hono-RNAi^JF02967attP2/+). (I) Similarly, TA increased the delay to the first spike in controls but not in the presence of yohimbine, in honoka mutants, or following honoka RNAi in MNs. (J–M) Firing responses of RP2 to ramp current injections in control (J) compared with Dmca1D RNAi in MNs (w*;OK371-GAL4/+; UAS-Dmca1D-RNAi^HMS00294attP2/+) (K). Representative current clamp traces before TA (Upper traces in J and K) during TA (Middle traces in J and K), and after washing (Lower traces in J and K) indicate that TA effects are reduced by Dmca1D RNAi in MNs. Quantification shows that Dmca1D RNAi in MNs abolishes the effects of TA on firing rate (L) and on the delay to the first spike (M). ***P < 0.001. n.s., not significant.

Fig. 4.

TA effects on crawling require the hono receptor and Dmca1D channels in MNs, and TBH levels are increased by starvation. (A) Representative crawling traces indicate that TA decreases crawling distance in CS controls (Upper traces) but not in honoka mutants (+;;oct-tyrR^hono/oct-tyrR^hono; Middle traces), or following honoka RNAi in MNs (w*;OK371-GAL4/+; UAS-hono-RNAi ^JF02967attP2/+; Lower traces). TA significantly decreases crawling speed (B) and distance (C) in controls but not in honoka mutants or following honoka RNAi in MNs. ***P < 0.001; Student’s t test. (D) Double immunolabeling for GFP (green) and TBH (magenta) in animals expressing GFP in TDC2-positive neurons (w;TDC2-GAL4/10xUAS-IVS-mcd8::GFP;+). (Di) TBH-positive puncta in TDC2 neuron somata. (Dii–Div) Z-projections of 2-µm thickness at different VNC depths (ventral, medial, and dorsal neuropils) reveal TBH positive puncta in central processes of TDC2 neurons. (E and F) Representative anti-TBH labeling in starved (E) and nonstarved (F) CS larvae. Starvation significantly increases TBH label in varicositylike processes (G) (Student’s t test; **P < 0.01) and in TDC2 neuron somata (H and I) (Student’s t test; **P < 0.01). (J) qRT-PCR reveals a significant increase in TBH mRNA but not of the housekeeping gene Rp49 following 2 h of starvation (**P < 0.01). (K) Sketch summary of the proposed OA/TA mechanisms that adjust locomotion to nutritional state. ICa²⁺, Ca²⁺ current; int., integrated; n.s., not significant; PNS, peripheral nervous system; SV, synaptic vesicle.