Modelling of BCS1L-related human mitochondrial disease in Drosophila melanogaster

Abstract

Mutations in BCS1L are the most frequent cause of human mitochondrial disease linked to complex III deficiency. Different forms of BCS1L-related diseases and more than 20 pathogenic alleles have been reported to date. Clinical symptoms are highly heterogenous, and multisystem involvement is often present, with liver and brain being the most frequently affected organs. BCS1L encodes a mitochondrial AAA + -family member with essential roles in the latest steps in the biogenesis of mitochondrial respiratory chain complex III. Since Bcs1 has been investigated mostly in yeast and mammals, its function in invertebrates remains largely unknown. Here, we describe the phenotypical, biochemical and metabolic consequences of Bcs1 genetic manipulation in Drosophila melanogaster. Our data demonstrate the fundamental role of Bcs1 in complex III biogenesis in invertebrates and provide novel, reliable models for BCS1L-related human mitochondrial diseases. These models recapitulate several features of the human disorders, collectively pointing to a crucial role of Bcs1 and, in turn, of complex III, in development, organismal fitness and physiology of several tissues.

Keywords: BCS1L; Drosophila melanogaster; Mitochondrial disease; Mitochondrial respiratory chain; Respiratory chain complex III.

Fig. 1

CG4908 is the D. melanogaster ortholog of human BCS1L. A Global alignment of fruitfly’s Bcs1 and human BCS1L proteins. Conserved domains are marked by coloured lines below the alignment, TMD by the blue line, internal MTS by the red line and AAA + ATPase domain by the yellow line. Mutated sites found in patients are indicated with stars above the sequence; Björnstad syndrome sites are marked by blue stars, complex III deficiency sites by yellow stars and the GRACILE syndrome site by a green star. The red dot shows the GRACILE mutation (S78G) modelled in mouse and the blue dot shows the yeast F342C mutation. B Subcellular localization of Drosophila Bcs1. Confocal micrographs showing Drosophila cells expressing Bcs1-HA (green immunofluorescence, Alexa Fluor 488), and immunolabelled with anti-ATP5A antibody to mark mitochondrial structures (purple immunofluorescence, Alexa Fluor 647). Sum intensity projections (z-stacks) are shown. Scale bars: 10 μm

Fig. 2

Ubiquitous KD of Bcs1 in D. melanogaster. A Relative percentage of egg to adult viability of ubiquitous Bcs1 KD cross (act5c-gal4 > CyO.GFP x UAS-shBcs1) and control cross (act5c-gal4 > CyO.GFP x w¹¹¹⁸), calculated at three developmental stages (eggs, pupae and adults), (n > 250, chi-square test 16.80 df(2), ***p < 0.001). B Mendelian frequencies of adults obtained by crossing heterozygous act5c-gal4 > CyO.GFP flies with homozygous UAS-shBcs1 flies (n = 172, chi-square test 123.7, df(1), ****p ≤ 0.0001). C Bcs1 expression levels normalized by Rp49 reference expression levels and measured by qPCR in whole larvae, expressed as relative quantity of template in the sample (RQ) in ubiquitous KD larvae (act5c-gal4 > UAS-shBcs1) compared to control larvae (act5c-gal4 > +). Data are plotted as mean ± S.D. (n = 3, Student’s t test ***p ≤ 0.001). D Morphological analysis of mouth hooks in whole-mount larval preparations. #act5c-gal4 > + control larvae; §UAS-shBcs1 > CyO.GFP control larvae; *act5c-gal4 > UAS-shBcs1 KD larvae. E Morphological evaluation of Bcs1 KD larvae. #act5c-gal4 > + control larvae; §UAS-shBcs1 > CyO.GFP control larvae; *act5c-gal4 > UAS-shBcs1 KD larvae. Scale bar: 1 mm. F Locomotor activity analysis of Bcs1 KD larvae. Right = peristaltic movements per minute in larvae; left = number of lines crossed per minute by larvae (on a 0.25 cm² grid). Genotypes: act5c-gal4 > + control larvae and act5c-gal4 > UAS-shBcs1 KD larvae (n = 3, Student’s t test ****p ≤ 0.0001)

Fig. 3

Biochemical characterization of Bcs1 KD larvae. A Enzymatic activities of MRC complexes (I–IV) were measured in parental control (act5c-gal4 > + , gray columns) and in ubiquitous KD larvae (act5c-gal4 > UAS-shBcs1, red column). Activities of MRC complexes were normalized to the activity of citrate synthase (CS). For each genotype, three biological replicates of mitochondrial preparations were analysed. Data are plotted as mean ± S.D. (n = 3, Student’s t test, **p ≤ 0.01, ****p ≤ 0.0001). B Blue-native gel electrophoresis (BNGE) analysis of MRC complexes in DDM-solubilized isolated mitochondria from control (act5c-gal4 > +) and Bcs1 KD larvae (act5c-gal4 > UAS-shBcs1). C BNGE, Western blot and immunodetection of DDM-solubilized mitochondria from Bcs1 KD and control flies with antibodies against a CIII subunit (UQCR-C2) and a CV subunit (ATP5A) as loading control. Bands were quantified by measuring the integrated density of the signal. Data are plotted as mean ± S.D. (n = 3, Student’s t test ***p ≤ 0.001). D Impl3 (Ldh homolog), Pfk (phosphofructokinase), Hex-A (hexokinase A) and PyK (pyruvate kinase) expression levels measured by qPCR in control (act5c-gal4 > +) and Bcs1 KD larvae (act5c-gal4 > UAS-shBcs1), expressed as relative quantity of template in the sample (RQ) compared to control. Data plotted are mean ± S.D. (n = 3, Student’s t test, *p ≤ 0.05; **p ≤ 0.01). E Quantification of total body lactate in control (act5c-gal4 > +) and Bcs1 KD larvae (act5c-gal4 > UAS-shBcs1), normalized to the protein content and expressed as relative quantity in the sample compared to control. Data are plotted as mean ± S.D. (n = 3, Student’s t test **p ≤ 0.01)

Fig. 4

Pan-neuronal KD of Bcs1 in D. melanogaster. A Bcs1 expression levels normalized by Rp49 reference expression levels and measured by qPCR in fly heads, expressed as relative quantity of template in the sample (RQ) in pan-neuronal KD larvae (elav-gal4 > UAS-shBcs1) compared to control larvae (elav-gal4 > +). Data are plotted as mean ± S.D. (n = 3, Student’s t test **p ≤ 0.01). B Morphological evaluation of pan-neuronal Bcs1 KD individuals hatching from the pupae. Scale bar: 1 mm. C Relative percentage of egg to adult viability of pan-neuronal Bcs1 KD cross (elav-gal4 x UAS-shBcs1) and control cross (elav-gal4 x w¹¹¹⁸), calculated at three developmental stages (eggs, pupae and adults), (n > 250, Chi-square test 6.747 df(2), *p < 0.05). D Representative survival curves (Kaplan–Meier) of pan-neuronal control (elav-gal4 > + , blue line) and KD (elav-gal4 > UAS-shBcs1, red line) flies under standard culture conditions. Statistical significance was tested by the log-rank (Mantel-Cox) test (***p ≤ 0.001)

Fig. 5

Neuromotor function in dopaminergic-specific Bcs1 KD flies. Climbing performance of dopaminergic control (TH-gal4 > +) and Bcs1 KD (TH-gal4 > UAS-shBcs1) female (red bars) and male (blue bars) flies at two different ages (2 and 10 days after hatching). Charts show mean and 95% CI, n = 60 animals. Statistical analysis was performed with one-way ANOVA with Dunn’s multiple comparisons test (*p ≤ 0.05, **p ≤ 0.01; ****p ≤ 0.0001)

Fig. 6

Muscle-specific KD of Bcs1 in D. melanogaster. A Bcs1 expression levels normalized by Rp49 reference expression levels and measured by qPCR in larval body wall muscles, expressed as relative quantity of template in the sample (RQ) in muscle-specific KD larvae (how24b-gal4 > UAS-shBcs1) compared to control larvae (how24b-gal4 > +). Data are plotted as mean ± S.D. (n = 3, Student’s t test **p ≤ 0.01). B Morphological evaluation of muscle-specific Bcs1 KD pupae. p = pupa, h = head, th = thorax, ab = abdomen, ey = eyes, l = legs, w = wings Scale bar: 1 mm. C Relative percentage of egg to adult viability of muscle-specific Bcs1 KD cross (how24b-gal4/TM3 x UAS-shBcs1) and control cross (how24b-gal4/TM3 x w¹¹¹⁸), calculated at three developmental stages (eggs, pupae and adults) (n > 250, Chi-square test 14.66 df(2), ***p < 0.001). D Mendelian frequencies of adults obtained by crossing homozygous how24b-gal4 flies with homozygous UAS-shBcs1 flies (n = 156, Chi-square test 111.9, df(1), **** p ≤ 0.0001).E Representative image of the progeny obtained by crossing homozygous how24b-gal4 flies with homozygous UAS-shBcs1 flies. Control flies (left vial) hatch from the pupa and reach the adult stage whereas muscle-specific Bcs1 KD flies (right vial) fail to hatch and die at the latest pupal stage

Fig. 7

Fat body–specific KD of Bcs1 in D. melanogaster. A Bcs1 expression levels normalized by Rp49 reference expression levels and measured by qPCR in adult abdominal fat body, expressed as relative quantity of template in the sample (RQ) in fat body–specific KD adults (ppl-gal4 > UAS-shBcs1) compared to control adults (ppl-gal4 > +). Data are plotted as mean ± S.D. (n = 3, Student’s t test ***p ≤ 0.001). B Relative percentage of egg to adult viability of fat body–specific Bcs1 KD cross (ppl-gal4 x UAS-shBcs1) and control cross (ppl-gal4 x w¹¹¹⁸), calculated at three developmental stages (eggs, pupae and adults), (n > 250, chi-square test 0.1039 df(2), ns = not significant. C Representative survival curves (Kaplan–Meier) of fat body control (ppl-gal4 > + , solid lines) and KD (ppl-gal4 > UAS-shBcs1, dotted lines) flies under standard culture conditions. Females and males are represented by red and blue lines, respectively. Statistical significance was tested by the log-rank (Mantel-Cox) test (****p ≤ 0.0001). D Quantification of TAG in control (ppl-gal4 > +) and Bcs1 KD abdominal fat body (ppl-gal4 > UAS-shBcs1), normalized to the protein content and expressed as relative quantity in the sample compared to control. Data are plotted as mean ± S.D. (n = 3, two-way ANOVA with Tukey’s multiple comparisons ns = not significant, *p ≤ 0.05, ****p ≤ 0.0001)