Larval and prepupal heat production and body temperatures.

(A) Representative output of real-time voltage changes in thermopiles of the small-sample chip calorimeter described by (Lerchner et al., 2024). Top graphs, signals obtained for Drosophila larvae, showcasing the noise attributed to movement (2σ). Quantitation of 2σ is shown in Supplementary Table S1. Bottom graphs, signals for prepupae; note that while control prepupae exhibit a stable heat output signal, 3XtubAOX prepupae show frequent heat bursts (black arrows), which are not attributed to mobility. Quantitation of heat burst events is also shown in Supplementary Table S1. Ref, a stable signal measured with an inert sample in a segment of the thermopile, used as reference. (B) Average heat production of larvae and prepupae at 25 or 13 °C, calculated as the voltage difference related to the Ref segment. Data represent means +/-standard deviation of 6-14 individuals. (C) Distance crawled and body wall contractions per min of control and 3XtubAOX larvae culture at 25 and 12 °C. Data represent means +/-standard deviation of 10-15 technical replicates from 10 biological replicates for each genotype. (D) Infrared thermographic measurements of changes in whole-body temperature over time in control (w1118) and AOX (3xtubAOX) larvae (top panel) and prepupae (bottom panel). Animals cultured at 25 °C were transferred to 12 °C and filmed using a thermographic camera, as described in the Materials and Methods. Each data point represents the body temperature of an individual animal from a biological replicate, and the curves represent the 1st order exponential fitting curve of temperature decay. (E) Representative infrared thermography images show control flies and AOX flies immediately after placement at 12 °C (time 0).

Larval biomass, size and food intake are increased by AOX.

(A, B) Total and relative masses, and (C) length and width of larvae, were measured as described in the Materials and Methods. Data are presented as means ± standard deviation (SD) from three biological replicates, each comprising at least 24 groups of 10 larvae. Significant differences are denoted as follows: “w1118 < 3XtubAOX”, differences between genotypes, and “25 °C < 12°C”, differences between temperatures. Letters ‘a–c’ indicate distinct statistical classes based on significant genotype-temperature interactions, determined by two-way ANOVA followed by Tukey’s post hoc tests (p < 0.05). (D) Representative light microscopy images of two wandering female L3 larvae cultured at 12°C. (E) Quantification of lipid droplets in fat body cells of wandering L3 larvae cultured at 25 °C, based on three biological replicates. Error bars represent ± SD of the mean; p < 0.05, based on unpaired t-test with Welch’s correction. (F) Lipid droplet area relative to total cell area. Data are presented as means ± SD from 4–5 independent samples. * indicates p < 0.05 according to Student’s t-test. (G) Representative confocal microscopy images of fat body cells showing lipid droplets. (H) Food intake, quantified as the percentage of the larval gut area stained with blue dye relative to the total L3 larval area (see Materials and Methods for details). Data represent means ± SD from 13-19 individuals sampled across two biological replicates. * denotes significant differences (p < 0.05), based on Student’s t-tests. (I) Representative light microscopy images of L3 larvae after ingesting dyed food.

Reconfiguration of the larval mitochondrial respiratory system in the presence of AOX.

Oxygen consumption rates (OCR, pmol / s * mg total protein) of wandering L3 larvae at 25 and 12 °C driven by Complex I (CI) and mitochondrial glycerol-3-phosphate dehydrogenase (mGPDH) were measured, respectively, after addition of pyruvate/malate and glycerol-3-phosphate/rotenone. The OXPHOS (A) and ET-capacity (B) states were achieved, respectively, after addition of ADP and a titration of the uncoupler carbonyl cyanide m-chlorophenylhydrazone. The Leak state (C) was obtained in separate assays following the addition of the ATP synthase inhibitor oligomycin. Data points represent the mean +/-standard deviation of 6 biological replicates. Letters ‘a–c’ indicate distinct statistical classes based on significant genotype X pathway(s) interactions, determined by two-way ANOVA followed by Tukey’s post hoc tests (p < 0.05). (D) OXPHOS inefficiency, calculated as the ratio Leak OCR / OXPHOS OCR, represents the OXPHOS coupling-control ratio (Gnaiger, 2020) and estimates the proportion of respiration not used in ATP production. (E) OXPHOS limitation, calculated as 1 – (OXPHOS OCR / ET-capacity OCR), expresses the limitation imposed on the electron transport system by the ADP phosphorylation system. Data points represent means +/-standard deviation of at least 4 biological replicates. * indicates significant differences (p < 0.05) according to Student’s t-tests. (F) Representative blue-native polyacrylamide gels showing the separation of mitochondrial complexes and supercomplexes from wandering L3 larvae (w1118, control line; AOX, 3XtubAOX line) grown at 25 °C and 12 °C. Mitochondria were solubilized with digitonin, and gels were either stained with Coomassie Blue or analyzed by in-gel activity. These analyses, combined with literature data, confirmed the positions of the indicated complexes. Labeled bands represent individual complexes (I-V), dimers (denoted by a subscripted “2”), or supercomplex structures. HMW-SC refers to high molecular weight supercomplexes. (G) Quantification of Coomassie-stained gels for the indicated complexes, dimers, or supercomplexes was performed via band densitometry. The density of each band was normalized to the density of the control at 25 °C within the corresponding lane. Data are presented as means ± standard deviation from two-three biological replicates. Letters ‘a–c’ indicate distinct statistical classes based on significant genotype X temperature interactions, determined by two-way ANOVA followed by Tukey’s post hoc tests (p < 0.05).

Mitochondrial uncoupling and complex I activity contributes to development of AOX-expressing flies.

Wet and dry masses of larvae of the indicated genotype cultured at 25 °C, with dietary supplementation of CCCP (A) or rotenone (E). Data represent means ± standard deviation (SD) from at least 8 replicates, each with 10 larvae. (B,C) Egg-to-pupa developmental time of control (w1118) and AOX-expressing (3XtubAOX) individuals at the indicated rotenone concentration and temperature. Letters ‘a-c’ indicate statistically distinct groups (p < 0.05), according to a two-way ANOVA with Tukey’s post hoc test. (D) Representative image showing lethality in w1118 larvae (dark individual larvae on top of diet) at 200 µM rotenone. (F) Representative fluorescence lifetime microscopy images of NADH obtained from fat body cells of wandering L3 larve at 25 °C, with the false-colored scales representing τ, ranging from shorter (red, free NADH) to longer lifetime (blue, enzyme-bound NADH). (G) Mean τm of free NADH autofluorescence of 5 different samples of larval fat bodies of each genotype. (H) Reverse amplitude rate (a2/a1) of enzyme-bound/free NADH, which provides estimates of the cellular NAD+/NADH ratio.

Targeted metabolomics indicates AOX-induced difference in the larval metabolism.

Levels of the indicated metabolites were determined by gas chromatography-mass spectrometry (GC-MS) to assess relative abundances among control (w1118) and AOX-expressing lines. The data was examined by principal component analyses (PCA) to show the overall differences between w1118 and 3XtubAOX L3 larvae (A), and among w1118, tubAOX35, tubAOX7 and 3XtubAOX mid-L2 larvae (D). The heat maps (B, E) and scatter plots (C, F) depict relative levels of the indicated metabolites. **, *** and **** indicate, respectively, significant differences with p < 0.01, p < 0.001 and p < 0.0001) determined by Student’s t-tests, and letters ‘a-c’ denote differences according to one-way ANOVA with Tukey’s post hoc test (p < 0.05). 2-HG, L-2-hydroxyglutarate; α-KG, α-ketoglutarate; G3P, glycerol-3-phosphate.

Effects of AOX expression on development.

(A) Relative transcript levels of AOX, Gpo1 and Ldh in wandering L3 larvae. Transcript levels were estimated using the levels of the housekeeping eIF-1A transcript as reference, and were normalized (arbitrarily set to 1.0) for AOX by its levels in AOX-expressing larvae (3xtubAOX) at 25 °C, and for Gpo1 and Ldh by their levels in control larvae (w1118) also at 25 °C. The data represent the average of two independent biological replicates, and the error bars standard deviations. Egg-to-pupa developmental time (B) and larval body masses (C) at the indicated temperatures for the background control w1118 line and the red-eye control UAS-empty2nd;UAS-empty3rd line, showing no effects of the white gene. (D) Representative results of in-gel activity for complex II (see Figure 3 for data quantitation). Larval body masses (E) and egg-to-pupa developmental time (F,G) of the w1118 and 3XtubAOX lines cultured on dietary supplementation with the uncoupler CCCP (E,F) and the CI inhibitor rotenone (G) at the indicated temperatures. Data represents the average of two-three biological replicates, each with at least five-ten technical replicates. * indicates significant differences according to Student’s t-test (p < 0.05). Two-way ANOVAs, followed by Tukey’s post hoc tests, were used to show differences (p < 0.05) between genotypes, between temperatures (denoted by the symbols < or =), or in the interaction genotype X temperature (denoted by the letters ‘a-c’).

AOX significantly contributes to larval respiration, particularly to the Leak state.

Remaining oxygen consumption rates (OCR) driven by the indicated dehydrogenase at the indicated temperature after inhibition of AOX with propyl gallate (A, C and E), or of CIII with antimycin A (B, D and F). The data represent percentages of the total OCR as shown in Figure 3A-C. The mitochondrial respiratory states OXPHOS (A and B), ET-capacity (C and D) and Leak (E and F) are shown. * indicates a significant difference (p < 0.05) between AOX and control larval samples, according to Student’s t-tests.

Metabolic analyses of AOX-expressing lines.

(A) Distribution of τm of free NADH autofluorescence (left graph) and of reverse amplitude rate of enzyme-bound/free NADH (a2/a1, right graph) of samples of larval fat bodies of each genotype. (B) Partial Least Squares Discriminant Analysis (PLS-DA) plots showing separation among the metabolic profiles of mid-L2 and wandering L3 larvae of the indicated genotypes, based on liquid chromatography-mass spectrometry. (C,D) Tendency of AOX dose-dependent increase in NAD+ and ATP levels, and in the NAD+/NADH and ATP/ADP ratios. (E) Heatmaps indicating the relative levels of the identified glycolytic and tricarboxylic acid cycle metabolites.

Larval agitation and prepupal heat burst events are increased by AOX.

The average noise of the voltage change output of the chip calorimeter (σ, see representative output in Figure 1C) +/-standard error represents larval movements inside the calorimeter chambers. The data was obtained from two biological replicates, using 6-10 individuals in each one (see details in Material and Methods). The average number of prepupal heat bursts per cycle and the percentage of cycles with heat bursts per individual were calculated for 10 individuals from each genotype in the ∼18 min cycle of each real-time measurement of the calorimetry assays.