(original chrisi) Complex interactions of octopamine and tyramine orchestrate sugar responsiveness and starvation resistance in Drosophila

Octopamine (OA) and its precursor tyramine (TA) are neurotransmitters operating in many different neuronal and physiological processes. We investigated the role of those two transmitters in Drosophila sugar responsiveness. Tyrosine-\(\beta\)-hydroxylase (t\(\beta\)h ) mutants are unable to convert TA into OA. Starved mutants show a reduced sugar response and their hemolymph sugar concentration is elevated compared to control flies. When starved to death, they survive longer. Temporally controlled rescue experiments revealed an action of the OA/TA-system during the sugar response, while spatially controlled rescue experiments suggest actions also outside of the nervous system. Additionally, the analysis of four OA- and four TA-receptor mutants suggests an involvement of both receptor types in the animals' physiological and neuronal response to starvation. These results complement the pharmacological investigations in Apis mellifera described in our companion paper (Buckemheimer et al.).


Fly stocks and culture
tßhnM18 (Monastirioti et al., 1996; FBal0061578), oamb (Han et al., 1998; OctαR, oamb286 FBti0038368, oamb584 FBti0038361), honoka (Kutsukake et al., 2000; Oct-TyrR, FBal0104701), hsp-tßh (Schwaerzel et al., 2003; FBal0152162) and w+;;UAS-tßh (Monastirioti, 2003; FBti0038601) were obtained from Henrike Scholz, Cologne; Hiromu Tanimoto, Martinsried; Andreas Thum, Konstanz; and Amita Seghal, Chevy Chase. Mutants that deleted large stretches of genomic DNA coding for the Octß2R (CG6989; FBgn0038063) were generated by remobilization of FRT-containing P-elements, i.e. f05679 and f07155 excised 17,539 bp giving rise to deletion mutant Octß2R3.22; f00021and f01537 excised 33,489 bp resulting in Octß2R4.3. Deletions were confirmed by genomic PCR. TyrRf05682 (CG7431f05682, FBal0184987), TyrRII29 (CG16766, FBgn0038541) and TyrRII-TyrR∆124were kindly provided prior to publication by Edward Blumenthal, Milwaukee (Zhang and Blumenthal, in preparation). Receptor mutants and their respective control lines were outcrossed for at least six generations into CS background.
Flies were kept on standard cornmeal/molasses-food in a 12/12 h light/dark cycle (light on at 8:00 hrs) at 60% relative humidity and 25°C except for hsp-tßhwhich were raised at 18°C without humidity control and except for flies used in electrophysiological experiments (see Electrophysiological recording).

Starvation procedure
Newly hatched to one day old flies were collected and transferred to fresh food vials. The following day, between 16:00 and 19:00 hrs, 20 to 30 flies of mixed gender were transferred into starvation vials (68 ml, Greiner bio-one, Frickenhausen, Germany) by a fly aspirator. The starvation vial contained a cotton pad moistened with 2.5 to 3 ml of Evian® water. If not otherwise indicated, starvation was performed at 25°C and 60% relative humidity and lasted for 20 h.

Survival experiments
Newly hatched to one day old flies were collected and transferred to fresh food vials. The following day, flies were briefly CO2-anesthetized and sorted by gender and genotype. At 17:00 hrs, around 20 female flies were transferred into a starvation vial (see Starvation procedure). Dead flies were counted every 3 h and not removed. Daily counting sessions were repeated from 9:00 to 18:00 hrs, until all flies were found dead.
For the survival rescue attempt, fly vials with eggs were stored in an incubator without humidity control and heated up every 4 h to 37°C for 30 min until hatching. Temperature in between heat shocks ranged from 25 to 30°C. Hatched flies were collected and transferred to fresh food vials and kept in the incubator with continuing heat shocks every 4 h.

Sugar response test
Newly hatched to one day old flies were collected and transferred to fresh food vials. The following day, they were starved as described (see Starvation procedure). Four hours before the end of the starvation period, female flies (if not stated otherwise) were briefly immobilized by cold-anesthesia. Their head and thorax were glued to a triangle-shaped copper hook (0.05 mm in diameter) using a UV sensitive glue (3M ESPE, Sinfony Indirect Lab Composite, Minneapolis, USA). Animals were then kept individually in small chambers (14 mm in diameter x 28 mm in height, custom-made) with ad libitum access to water until the test. hsp-tßh flies were glued before receiving the heat shock.
Tests were performed between 12:00 and 16:00 hrs. Using forceps, we transferred flies by their hook and fixed them to a magnetic clamp, which was then attached to a rack. This treatment established free movement of the flies’ tarsi and proboscis and was a modification from a previously described PER assay (Scheiner et al., 2004) in order to prevent unnecessary stress and pressure on the abdomen of the flies. A group of six to eight flies was tested in parallel. A filter paper soaked with sucrose solution was presented for 5 s to all six tarsi but not the proboscis. Seven different concentrations (0, 0.1%, 0.3%, 0.6%, 1%, 3%, and 30%) were presented in series with an inter-stimulus interval of 80 s. The proboscis extension response was recorded. Finally, the proboscis was stimulated by 30% sucrose solution. Flies not responding to this last stimulation or responding to the first stimulation (water only) were discarded from the analysis.
For the first sugar response rescue attempt (Fig. 1A), flies were starved at 18°C and put into an incubator without humidity control and heated up to 37°C for 30 to 45 min. After the heat shock, flies were kept in a 25°C incubator with humidity control for 3 h until testing. For the second rescue attempt (Fig. 1B), the first heat shock was given with beginning of starvation every 23 h for 45 min until one day before testing. Temperature in between heat shocks was 18°C.

PER assay validation
As a positive control for the locomotion-independent sugar response assay, we tested the sugar response of wild type flies starved for 14 h versus 21 h (Colomb et al., 2009). There was a statistically significant increase in sugar response with the longer starvation regime (p = 0.0409, Wilcoxon rank sum test, n = 48; Damrau et al., 2014). We thus conclude that the assay is sensitive enough for detecting slight differences in sugar responsiveness.

Carbohydrate measurement
Newly hatched to one day old flies were collected and transferred to fresh food vials. The following day at 17:00 hrs, 20 flies of mixed gender were either transferred into starvation vials (see Starvation procedure) or kept in the food vials. After 20 h, approximately 40 female flies per group were cold-anesthetized, pierced through the thorax by the tip of a dissecting needle (0.5 mm in diameter), and collected on ice within a sieve composed of two tubes. The hemolymph was centrifuged out of the fly into the bottom tube at 4°C. 0.5 µl of the extracted hemolymph was transferred by a capillary (0.5 µl, Hirschmann Laborgeraete, Eberstadt, Germany) into 19.5 µl PBS (see Damrau et al., 2013). Trehalose and glucose content in the hemolymph was measured according to the protocols provided by the manufacturer (Sigma Aldrich, Seelze, Germany). 10 µl of the hemolymph-PBS mixture were added to 30 μl citric acid buffer (135 mM, pH 5.7 at 37°C) and 10 μl of a trehalase enzyme solution (Sigma Aldrich, 3% in citric acid buffer). After incubation overnight at 37°C, 50 μl of Tris buffer were added. 80 µl of the resulting solution were added to 156.8 μl Glucose oxidase and 3.2 μl o-Dianisidine (Glucose Assay Kit, Sigma Aldrich) and incubated for 30 min at 37°C. Finally, 160 μl of 33% sulfuric acid were added. Absorbance at 540nm was measured for the resulting solution using a nanoDrop® (nanoDrop Technologies, Wilmington, USA) spectrometer.

Electrophysiological recording
Flies were raised on cornmeal-yeast-glucose-agar medium under a 12/12 h light/dark cycle (lights on at 06:00 hrs) at 25°C. Newly hatched to one day old flies were collected and transferred into a vial containing Kimwipe paper soaked with 100 mM glucose for one to two days as previously described (Zhang et al., 2010). Starved flies were kept in a vial containing Kimwipe paper soaked with Evian® water for 20 h before testing.
Electrophysiological recordings from l-type labellar chemosensilla were done by the tip-recording method, as previously described (Hodgson et al., 1955; Hiroi et al., 2002). Briefly, the proboscis was fixed at the base of the labellum. A glass capillary filled with Drosophila Ringer solution served as an indifferent electrode. 100 mM sucrose solution for stimulation contained 1 mM KCl as electrolyte. The recorded signals were digitized and analyzed using the custom software dbWave (Marion-Poll, 1995, 1996). Action potentials were detected by a visually-adjusted threshold set across the digitally filtered signal. The total number of spikes within 1 second was counted.

Figures and statistical analyses were performed in R (http://r-project.org). If not stated otherwise, data are illustrated as boxplots representing the median (line), the 25% and 75% quartiles (boxes), the data within 1.5 times the interquartile range (whiskers), and data outside that range (outliers, depicted as points).
The sugar response score was calculated as the sum of all positive responses over the seven sucrose presentations and therefore ranges from 0 to 7 (Total PER).
For survival measurement, the mean proportion of animals still alive was calculated over time. The LD50 for each tested group, i.e. the time point at which 50% of the flies were dead, was estimated using MASS-package in R.
Hemolymph carbohydrate content was read from a calibration curve showing the absorbance of standard glucose/trehalose solutions that were treated identically to hemolymph. Change in sugar content was calculated as
(intensitystarved - intensityfed) / (intensitystarved + intensityfed).
The significance level of statistical tests was set to 0.05. Depending on data distribution, we used parametric (Welch Two-Sample t-Test or two-way ANOVA followed by TukeyHSD post hoc test) or non-parametric (Wilcoxon rank sum test or paired Wilcoxon rank sum test with Bonferroni correction) tests as depicted in the figure legends.