NDUFB7 mutations cause brain neuronal defects, lactic acidosis, and mitochondrial dysfunction in humans and zebrafish

Abstract

Complex I of the mitochondrial electron transfer chain is one of the largest membrane protein assemblies ever discovered. A patient carrying a homozygous NDUFB7 intronic mutation died within two months after birth due to cardiorespiratory defects, preventing further study. Here, we report another patient with compound heterozygous mutations in NDUFB7 who suffers from pons abnormality, lactic acidosis, prematurity, prenatal and postnatal growth deficiency, incomplete closure of the abdominal wall (ventral hernia), and a poorly functioning gastrointestinal tract (pseudo-obstruction). We demonstrated that the patient's skin fibroblasts are deficient in Complex I assembly and reduced supercomplex formation. This report further broadens the spectrum of mitochondrial disorders. The patient has had several surgeries. After receiving treatment with Coenzyme Q10 and vitamin B complex, she has remained stable up to this point. To further explore the functionality of NDUFB7 in vivo, we knocked down Ndufb7 in zebrafish embryos. This resulted in brain ventricle and neuronal defects, elevated lactic acid levels, and reduced oxygen consumption, indicating defective mitochondrial respiration. These phenotypes can be specifically rescued by ectopic expression of ndufb7. More importantly, Mitoquinone mesylate (MitoQ), a common remedy for mitochondrial disorders, can ameliorate these conditions. These results suggest a role for NDUFB7 in mitochondrial activity and the suitability of the zebrafish model for further drug screening and the development of therapeutic strategies for this rare disease.

Fig. 1. NDUFB7 is the disease causative gene.

A The patient’s brain MRI T2-weighted sagittal (left) or axial image obtained at 4 years old reveals high-intensity pons lesions (gray and pointed by arrows). B Partial Sanger sequencing chromatographs of the patient and parents. The left panel shows the presence of a deletion (denoted by orange bars) in the patient and her mother, while the right panel shows a missense mutation (denoted by orange arrows) found in the patient and her father. C The upper panel shows the gene structure of the human NDUFB7 gene (NM_004146.5, 456 nucleotides (nt)), which has three exons and two introns (the intron between Exon 2 and 3 is indicated). The orange lines indicate two mutations in Exon 2 and 3 of the patient’s NDUFB7 gene. The NDUFB gene is translated to 137 amino acids (aa, NP_004137.2, a cartoon in the middle panel). A large portion of Exon 2 and 3 encodes a coiled-coil-helix-coiled-coil-helix (CHCH) domain marked in orange. The lower panel presents a sequence alignment of sequences comprising the mutations and the CHCH domains among human, mouse, dog, and zebrafish (“*” identical; “:” similar). D Respiratory chain (RC) activities in cultured fibroblasts from the patient expressed as a ratio (U/U) relative to citrate synthase (CS) activity. The patient’s cells exhibited higher Complex II activity but lower activities in other complexes.

Fig. 2. Mitochondrial complex formation is disrupted in fibroblasts derived from the patient carrying the NDUFB7 mutations.

Using blue native polyacrylamide gel electrophoresis (BN-PAGE), mitochondrial extracts from control (C) and patient (P) skin fibroblasts were separated using Triton A X-100 or B digitonin. These extracts were probed for the indicated protein in parentheses to identify Complexes I–IV. Complex II served as a loading control, as shown at the bottom of each well, except in the Complex II lane. C The samples were further separated by 2D-SDS PAGE and probed for the indicated mitochondrial accessory proteins.

Fig. 3. Knockdown of Ndufb7 changes the sizes of brain ventricles.

We microinjected 1-cell stage zebrafish embryos without (untreated) or with indicated reagents and photographed them under a stereomicroscope at 48 hours post-fertilization. A Representative images for each treatment are presented. The ventricles are enclosed in red and white dotted lines for the midbrain and hindbrain, respectively. The area of the ventricle was measured in pixels using the Image J software. All data are shown in a scatter plot with mean ± standard error of the mean (SEM) for each treatment shown at the bottom of each column for B midbrain and C hindbrain, respectively, from three independent experiments. **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 4. Knockdown of Ndufb7 changes the sizes of brain ventricles revealed by dextran rhodamine injection.

A Zebrafish embryos were treated as described in Fig. 3, cultured until 48 hours post-fertilization, injected with dextran rhodamine into their brain ventricles, and photographed under a black field using a rhodamine cube. Representative images are presented in lateral view. B, C The midbrain (smaller chamber at the anterior) and hindbrain (larger chamber at the posterior) are enclosed by dotted lines. The area of the ventricle was measured in pixels, analyzed, and shown as described in Fig. 3 from three independent experiments. NS: not significant; **p < 0.01; ***p < 0.001.

Fig. 5. Knockdown of Ndufb7 reduces the neuronal volume of the brain.

Transgenic Tg(Huc:kaede) zebrafish embryos were treated as described in Fig. 3, cultured until 48 hours post-fertilization, observed using a 10X objective, and photographed under a black field using a GFP cube under confocal microscopy. A All images are presented in lateral view. B The midbrain (mb) and hindbrain (hb) are pseudocolored in yellow and white, subjected to volume measurement using the Metamorph software, as shown in C midbrain and D hindbrain. The volume of each brain compartment was measured in voxels. Data from three independent experiments are presented, analyzed, and shown as described in Fig. 3. NS not significant; **p < 0.01; ***p < 0.001.

Fig. 6. Knockdown of Ndufb7 increases the lactic acid in zebrafish embryos.

We injected 1-cell stage zebrafish embryos without or with indicated reagents as described in Fig. 3, cultured to 24 h post fertilization, lyzed, and subjected to lactic acid measurement. Data from four independent experiments are presented, analyzed, and shown as described in Fig. 3. NS not significant; **p < 0.01.

Fig. 7. Knockdown of Ndufb7 reduces the oxygen consumption rate.

We treated one-cell stage zebrafish embryos without or with indicated Ndufb7 translational-blocking morpholino oligonucleotides (tMO), and Ndufb7 mRNA or Mitoquinone mesylate (MitoQ) as described in Fig. 3, cultured to 24 hours post-fertilization, and subjected them to the measurement of oxygen consumption rate (OCR) using the Seahorse XFe24. A Graphical depiction illustrating the changes in mitochondrial respiration upon exposure to oligomycin (18.7 μM), carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP; 8 μM), and rotenone/antimycin A (3.5 μM). In 203 min for each measurement cycle, each inhibitor was injected at a predetermined time point as indicated by a dotted line. The blue, orange, gray, and yellow curves represent the untreated, tMO-injected embryos and tMO with Ndufb7 RNA or MitoQ, respectively. Calculations were made to show B Basal respiration, C maximal respiration, and D adenosine triphosphate (ATP) production. The error bar indicates the standard deviation. Data from three independent experiments are presented, analyzed, and shown as described in Fig. 3. NS not significant; **p < 0.01.