We have provided a more detailed protocol for the Pharyngeal Pumping Assay protocol for Drosophila Melanogaster, as detailed in our publication. Specifically, we have added some additional information about positioning the preparations for Video Recording, and included specific details about how to conduct Data Analysis on the videos.

Preparation of Animals:

We employed a pharyngeal pumping assay, modified from that of Manzo et al. (2012). Mated females were aged for 10–14 days to allow for tetanus toxin (TNT) or Chrimson (CHR) to accumulate via expression from the relatively weak IR60b-GAL4 driver. Flies were also aged for 10–14 days in IR60b¹, IR60b², and IR94f-GAL4 experiments. Before testing, flies were placed in a 1000-μl pipette tip. A second 1000-μl pipette tip was inserted into the first tip so as to contain the fly for a starvation period of 12–14 hr. Flies were placed in a 100-mm Petri dish with three Kimwipes wetted with 5 ml water, and then transferred to a humidity-controlled room to prevent dehydration.

Preparation of Pharyngeal Pumping Assay:

After starvation, the fly was aspirated into another 1000-μl pipette tip such that it was immobilized with its head and proboscis exposed. Aspiration involved attaching plastic tubing to the end of the 1000-μl pipette, and having the investigator gently blow air into the tip. The fly would then be gently pushed into the narrow point at the end of the 1000-μl pipette, until its head was pushed through the narrow point. Once through, the head was now positioned in the open air on the other side of the tip, while the thorax was sufficiently restrained to immobilize the fly without harming it. A second 1000-μl pipette was inserted into the large end of the first 1000-μl pipette, to prevent the fly from escaping, if it managed to move backwards, away from the narrow point where it had originally been restrained. Such “escapees” would be re-aspirated into the tip, if possible. Occasionally, ends of the 1000-μl pipette tips were trimmed with a razor blade to widen the opening for the fly’s head. Flies were then mounted on a micromanipulator. A clamp accessory was attached to a micromanipulator, which was then closed on the larger end of the 1000-μl pipette tip. The fine controls were then used so the fly could be brought into the field of view under a dissecting microscope, such that the underside of its proboscis could be viewed, including the cibarium. For our experiments, video recording was performed by using a Nikon SMZ800 dissecting microscope with an attached Sony HD Camcorder, which was in turn hooked up to a computer so real time imaging/recording could occur during feeding. Before food was delivered, the fly was presented a water droplet to ensure the animal was not overly desiccated from starvation. If it consumed water for longer than 10 s, the animal was discarded. Only ~5% of flies were discarded by this criterion.

We added 0.4 μg/μl erioglaucine blue dye to the liquid food to facilitate data acquisition. Food was presented with a P20 PipetteMan mounted on a micromanipulator, allowing for fine adjustments during delivery. The P20 PipetteMan was also mounted to the micromanipulator, using a clamp adapter. To bring the food to the fly, 20.0 μl was loaded into the P20 PipetteMan. The knob was the slightly rotated, to bring the volume of the P20 PipetteMan to 19.7 μl. This would be sufficient to “push” a droplet roughly 0.3 μl in size from the pipette tip, for the fly to sample. The knob of the P20 would be slightly rotated in 0.3 μl increments for each subsequent fly, until the food needed to be reloaded in the P20.

During video recordings, the fly was first brought into position, such that you could see the underside of its proboscis, with the camera focusing on it’s cibarium. From this point, the fly was left stationary on its micromanipulator. Then, the food drop was brought to the fly using the micromanipulator that moved the P20 PipetteMan, using small adjustments to move the droplet to the labellum of the fly. Video was left recording throughout the feeding process.

The fly was offered food for 2 s. Flies that ingested liquid were allowed to continue feeding until they freely terminated feeding. After breaking contact with the drop, flies were given 3 s of rest before a subsequent presentation, in which they were given another 2 s to initiate a second bout. This process was repeated until the fly no longer responded to food. Typically, 2 s is sufficient presentation time to initiate feeding, as longer presentations did not increase the likelihood of initiating a feeding bout. On average, flies engaged in 1–3 bouts. If a fly did not initiate feeding after four attempted presentations, it was discarded as a non-responder. When testing different concentrations and tastants, the fraction of non-responders did not differ significantly between control, mutant, or transgenically manipulated flies, indicating there were no differences in the capacity to initiate feeding between the different genotypes. Typically, the fraction of non-responders was less than ~10% for appetitive sugars like sucrose.

Each fly was used for only one experiment to prevent previous experience from influencing its responses. The investigator was blind to the genotypes of the flies. The video was analyzed using QuickTime Media Player. Within each experiment, responses of different genotypes were measured in parallel, at the same time of day, by the same investigator, and within the same 3–5 day time window.

Data Analysis:

We analyzed the videos for different feeding parameters, including (1) Total Feeding Time, (2) Swallowing Rate, (3) Volume Ingested.

Total Feeding Time: Total Length of time was calculated by reviewing the videos in QuickTime Media Player, and summing up the length of time flies spent actively drinking the droplet of food. Times were obtained by marking specific time-stamps from when feeding bouts began and when feeding bouts ended, and calculating the difference between the two. In flies that had multiple bouts, times for all the separate bouts were added together to get a Total Feeding Time value for that fly.

Swallowing Rate: To see if Total Feeding Time correlated linearly with Swallowing Rate, and to test if Swallowing Rate remained constant across all Sucrose concentrations, we directly quantitates fly “swallowing” during feeding, i.e. pumping their cibarium in the pharynx. To quantitate fly swallowing throughout the experiment, we converted the video to a grayscale file, which has a pixel intensity range from 0 to 256. Then, using Image J, we analyzed videos by drawing an elliptical Region of Interest (ROI) around the fly’s cibarium, which is easily detected due to the darker-shaded dye. Then, using the function in ImageJ that measured pixel intensity in the ROI over the course of a video timelapse, we would plot image intensity over time for the course of the experiment. Briefly, when the cibarium is full of food, there a large amount of blue dye in the pharynx, which scans as higher intensity, and thus a higher value on the 0 to 256 grayscale index. When the cibarium empties as the fly swallows, there is little blue dye in the pharynx, which scans as higher intensity, and thus a lower value on the 0 to 256 grayscale index. Therefore, plotting ROI pixel intensity vs. time would yield a series of peaks while the fly repeatedly swallowed during feeding bouts. We would then use the Analyzer function in ImageJ the count the number of peaks during the feeding sessions, and then divide the total number of peaks by the duration of the feeding session time to get the Swallowing Rate. We found that Swallowing Rate was consistent across all sucrose concentrations, and that Swallowing Rate also correlated linearly with Time Total Feeding (see Figure 3 for Statistical Analyses).

Volume Ingested: We also directly measured how much food flies ate after feeding sessions. Volumes were determined by extracting dye from single flies after feeding, and measuring the absorbance of each sample. Absorbance values were converted into calculated volumes using the slope of a standard absorbance curve for the concentration of the blue dye. Specifically, volumes ingested by single flies were measured as follows: 10 serial two-fold dilutions of 0.4 μg/μl erioglaucine were prepared for spectrophotometer analysis using a NanoDrop 2000c Spectrometer (Thermo Scientific: Wilmington, DE). The average optical density (OD₆₃₀) was determined for 1 μl of each dilution (n ≥ 3 for each concentration). Average ODs were then used to plot a standard line for absorbance of blue dye (OD) versus the dye concentration. The resulting slope was 0.14 ± 0.001 OD/μl, and was used as a factor for converting OD values of 0.4 μg/μl erioglaucine to volume of fluid ingested by the fly. A single mated female was fed 900 mM sucrose with 0.4 μg/μl erioglaucine blue dye in the pharyngeal pumping assay. After feeding, the fly was homogenized in water and samples were centrifuged for 1 min at 14,000 RPM to pellet the cuticle. 1 μl of supernatant from homogenized flies were then assayed with the NanoDrop Spectrometer, and an average OD value was used to determine the amount of blue dye present in each fly (n ≥ 4 OD values were averaged for each fly). The slope of the standard line for 0.4 μg/μl erioglaucine was then used to convert average OD values into volumes ingested. To control for endogenous absorbance of the extract, four females were fed 900 mM sucrose without blue dye, and OD values for each fly were obtained. The calculated values for these control flies were averaged, yielding a standard absorbance value in the OD₆₃₀ range. This standard value was subtracted from each measurement from the flies that were fed 900 mM sucrose with 0.4 μg/μl erioglaucine, before determining the final calculated ingested volumes.

Reference:

Manzo A, Silies M, Gohl DM, Scott K. Motor neurons controlling fluid ingestion in Drosophila. PNAS. 2012;109:6307–6312. doi: 10.1073/pnas.1120305109