Feeding difficulties, a key feature of the Drosophila NDUFS4 mitochondrial disease model

Abstract

Mitochondrial diseases are associated with a wide variety of clinical symptoms and variable degrees of severity. Patients with such diseases generally have a poor prognosis and often an early fatal disease outcome. With an incidence of 1 in 5000 live births and no curative treatments available, relevant animal models to evaluate new therapeutic regimes for mitochondrial diseases are urgently needed. By knocking down ND-18, the unique Drosophila ortholog of NDUFS4, an accessory subunit of the NADH:ubiquinone oxidoreductase (Complex I), we developed and characterized several dNDUFS4 models that recapitulate key features of mitochondrial disease. Like in humans, the dNDUFS4 KD flies display severe feeding difficulties, an aspect of mitochondrial disorders that has so far been largely ignored in animal models. The impact of this finding, and an approach to overcome it, will be discussed in the context of interpreting disease model characterization and intervention studies.This article has an associated First Person interview with the first author of the paper.

Keywords: Disease model; Drosophila model; Feeding impairment; Mitochondrial disease.

Fig. 1.

Morphofunctional characterization of mitochondria in ubiquitous dNDUFS4 KD flies. (A) Enzymatic activity measurement of the mitochondrial OxPhos complexes normalized on citrate synthase (CS). CI activity was reduced to 18.1%±2.8% (***P<0.0001) and CIII activity was increased to 140.4%±23.7% (*P<0.0102) in ubiquitous KD flies (mean± s.e.m., n=5, two-way ANOVA). (B-G) Transmission electron microscopy of indirect flight muscle. (B) Representative image of mitochondria in controls. (C) Representative image of mitochondria in dNDUFS4 KD flies. Mitochondria with dispersed cristae are indicated by black arrowheads. In D-G, asterisks indicate examples of mitochondria with structural defects found in ubiquitous KD fly samples. (D) Mitochondrion with onion-like inner membrane. (E) Swollen mitochondrion. (F) Mitochondrion with sparse matrix/distant cristae. (G) Mitochondria with irregular cristae densities. Scale bars: 1 µm in B and C; 2 µm in D-G.

Fig. 2.

The ubiquitous dNDUFS4 KD model presents feeding deficits. Control flies are presented on the left and ubiquitous KD flies on the right, as indicated. (A) Representative images of abdomen evolution over time upon regular cornmeal feeding, highlighting a progressive shrinkage of the KD fly abdomen. All dNDUFS4 KD flies presented abdominal shrinkage. D0 represents the day of collection (0 to 1 day old), D1 represents 1-2 days old, etc. (B,C) Analysis of the digestive system upon solid and liquid food supplemented with 0.5% Bromophenol Blue. Upon regular cornmeal fly food, the digestive system of all ubiquitous KD flies appeared blue-free (B). When subjected to 5% sucrose with Bromophenol Blue delivered in capillaries, both control and ubiquitous dNDUFS4 KD flies were able to ingest liquid food formulation, although the KD flies seemed to eat qualitatively less (C). White arrowheads indicate the crop and gut. (D) Effects of liquid feeding on progressive abdomen shrinkage. The KD flies with liquid feeding presented a dramatically slower shrinkage compared with the solid-feeding flies (note the different last timepoint of the time series).

Fig. 3.

Effects of solid and liquid feeding on the lifespan of control and ubiquitous dNDUFS4 KD flies. Survival curve of ubiquitous dNDUFS4 KD flies and their corresponding controls fed with regular cornmeal medium or 5% sucrose. Solid-fed ubiquitous dNDUFS4 KD flies (n=198) die significantly faster than their genetic controls (n=222), with median survival (day at which 50% of the flies have died) at day 3 for KD flies and day 57 for control flies. When fed with 5% sucrose in capillaries, ubiquitous KD flies (n=325) displayed a reduced lifespan compared with control flies (n=249), with respective median survival at day 7 and day 34 (Kaplan–Meier curve, log-rank Mantel–Cox test, ****P<0.0001).

Fig. 4.

Ubiquitous dNDUFS4 KD flies display locomotion impairments. (A) The spontaneous locomotion assay highlighted a reduced total walked distance of solid-fed ubiquitous KD flies (905.6±105 mm/7 min; n=19) compared with controls (2248±185.3 mm/7 min; n=20). The liquid-fed dNDUFS4 KD flies had a severely reduced total walked distance over 7 min of acquisition (220.9±49.9 mm/7 min; n=18) compared with the controls (2441±202.7 mm/7 min; n=20) and their solid-fed siblings (***P<0.0001, unpaired Student's t-test, mean±s.e.m.; ns, nonsignificant). (B) Climbing abilities or time necessary for 70% of the flies to climb 9.5 cm. The climbing assay revealed impaired negative geotaxis of the solid-fed ubiquitous KD flies (30.7±1.920, n=8) compared with controls (5.2±0.2, n=4) (***P<0.0001). The liquid-fed ubiquitous dNDUFS4 KD flies also presented with negative geotaxis impairment (31.6±1.3 s, n=10) compared with the controls (5.3±0.2, n=4) (***P<0.0001, unpaired Student's t-test, mean±s.e.m.; ns, nonsignificant). Diet did not improve or impair the climbing abilities of ubiquitous KD flies (P<0.67). (C) Island assay. Solid-fed ubiquitous KD flies spent significantly (***P<0.0001) more time on the platform as expressed by the area under the curve (AUC) (685.2±156.7, n=8, for the controls; 6028±247.6, n=13, for the ubiquitous KD flies). The liquid-fed ubiquitous dNDUFS4 KD flies also spent significantly more time on the platform (AUC, 852.3±231.1; n=8) than the control flies (4727±584.2; n=8) (***P<0.0001). Liquid feeding improved the performances of ubiquitous KD flies in the island assay (*P<0.0293) (unpaired Student's t-test, mean±s.e.m.; ns, nonsignificant).

Fig. 5.

Ubiquitous dNDUFS4 KD flies have a strongly reduced pericerebral fat body. (A,B) Transverse sections of 1-day-old fly heads stained with Toluidine Blue. Black arrows point at the pericerebral fat body. The intensity of the histological staining, Toluidine Blue, is not indicative of differences in fat composition. (A) Control and ubiquitous KD flies fed with solid food. (B) Control and ubiquitous KD flies fed with liquid food. Both solid- and liquid-fed KD flies displayed dramatic pericerebral fat body depletion.

Fig. 6.

Characterization of muscle- and neuron-specific NDUFS4 KD. (A) Survival curve of flies collected at D0 fed with regular cornmeal fly medium. Muscle KD flies (n=151) die significantly faster than their genetic controls (n=85) (****P<0.0001), with a median survival of 11 and 49 days, respectively. (B) Abdomen photography of flies fed with regular cornmeal medium supplemented with 0.5% Bromophenol Blue highlighted important feeding difficulties, although the intestines showed a faint blue coloration. (C) The neuronal KD flies (n=167, median survival 45 days) died at an earlier age than their isogenic controls (n=129, median survival 58 days) (****P<0.0001) (Kaplan–Meier curve, log-rank Mantel–Cox test, GraphPad Prism 6). (D) Food supplemented with 0.5% Bromophenol Blue also highlighted feeding difficulties, but with a milder degree of severity. White arrowheads indicate the crop and gut.