Exclusion of sulfide:quinone oxidoreductase from mitochondria causes Leigh-like disease in mice by impairing sulfide metabolism

Abstract

Leigh syndrome is the most common inherited mitochondrial disease in children and is often fatal within the first few years of life. In 2020, mutations in the gene encoding sulfide:quinone oxidoreductase (SQOR), a mitochondrial protein, were identified as a cause of Leigh syndrome. Here, we report that mice with a mutation in the gene encoding SQOR (SqorΔN/ΔN mice), which prevented SQOR from entering mitochondria, had clinical and pathological manifestations of Leigh syndrome. SqorΔN/ΔN mice had increased blood lactate levels that were associated with markedly decreased complex IV activity and increased hydrogen sulfide (H2S) levels. Because H2S is produced by both gut microbiota and host tissue, we tested whether metronidazole (a broad-spectrum antibiotic) or a sulfur-restricted diet rescues SqorΔN/ΔN mice from developing Leigh syndrome. Daily treatment with metronidazole alleviated increased H2S levels, normalized complex IV activity and blood lactate levels, and prolonged the survival of SqorΔN/ΔN mice. Similarly, a sulfur-restricted diet normalized blood lactate levels and inhibited the development of Leigh syndrome. Taken together, these observations suggest that mitochondrial SQOR is essential to prevent systemic accumulation of H2S. Metronidazole administration and a sulfur-restricted diet may be therapeutic approaches to treatment of patients with Leigh syndrome caused by mutations in SQOR.

Keywords: Metabolism; Mitochondria; Mouse models; Neurological disorders; Therapeutics.

Figure 1. Sqor^ΔN/ΔN mice had decreased weight gain, progressive hypothermia, motor dysfunction, and shortened lifespan.

(A) Kaplan-Meier survival probability curve for Sqor^ΔN/ΔN and control mice. Log-rank P value and sample size are shown. (B) Body weight trajectory of Sqor^ΔN/ΔN and control mice. n = 10 mice for each group. Data are presented as means with SD. (C) Body temperature trajectory of Sqor^ΔN/ΔN and control mice. n = 10 mice for each group. Data are presented as means with SD. (D) Time to fall from rotarod in Sqor^ΔN/ΔN and control mice at postnatal ages 20, 30, 40, and 50 days. Comparisons were made using unpaired 2-tailed t test. n = 4 to 10 mice for each group. Data are presented as means with SD.

Figure 2. Sqor^ΔN/ΔN mice had decreased complex IV activity, increased systemic H₂S levels, and an impaired metabolic status.

(A) Histochemical staining was used to measure cytochrome c oxidase (COX) activity in Sqor^ΔN/ΔN and control mice. Control mouse brain, liver, and muscle sections are stained dark brown; the same tissues from Sqor^ΔN/ΔN mice are light brown, indicating decreased complex IV activity. Scale bars: 100 μm. (B) Complex IV activity in isolated mitochondria of the brain, liver, and muscle. The mean values of complex IV activity in control mice were set to 100%. Comparisons were made using unpaired 2-tailed t test. n = 5–6 mice for each group. Data are presented as means with SD. (C) Sulfide levels in plasma, brain, liver, and muscle were measured using HSip-1. The sulfide levels in Sqor^ΔN/ΔN mice were compared with those in control mice (the sulfide level for control mice was set to 1). Data were analyzed using Mann-Whitney test for plasma, and unpaired 2-tailed t test for brain, liver, and muscle. n = 5–8 mice for each group. Data are presented as medians with interquartile range for plasma, and as means with SD for brain, liver, and muscle. (D) Compared with control mice, the blood lactate levels in Sqor^ΔN/ΔN mice were higher at postnatal ages 30, 40, and 50 days. Comparisons between Sqor^ΔN/ΔN and control mice at postnatal ages 30, 40, and 50 days were made using unpaired 2-tailed t test. Mixed-effects analysis with Dunnett’s multiple-comparison test was performed to compare the blood lactate levels at postnatal ages 30, 40, and 50 days in Sqor^ΔN/ΔN mice. n = 4–5 mice for each group. Data are presented as means with SD.

Figure 3. Sqor^ΔN/ΔN mice had lesions in basal ganglia, midbrain, and brain stem.

T2-weighted brain MRI was performed on three 8-week-old Sqor^ΔN/ΔN mice: Sqor^ΔN/ΔN mouse no. 1 (A), Sqor^ΔN/ΔN mouse no. 2 (B), and Sqor^ΔN/ΔN mouse no. 3 (C). Brain lesions in Sqor^ΔN/ΔN mice are indicated by yellow arrows. Brain MRI scans of a control mouse, at the same anatomical levels as those of Sqor^ΔN/ΔN mice, are shown as reference in the bottom panels of A, B, and C.

Figure 4. Sqor^ΔN/ΔN mice had neurodegeneration with gliosis, acute and subacute hemorrhage, and brain tissue hyperoxia.

(A) The portions of the inferior colliculus of Sqor^ΔN/ΔN mouse no. 1 that were indicated in the brain MRI scans were used for immunohistochemical staining (upper panels). Staining for NeuN, Iba-1, and GFAP was performed to evaluate neurodegeneration and gliosis. The corresponding portions of the inferior colliculus of a control mouse were used for control staining (lower panels). An enlargement of the boxed regions is provided below each cross section. Scale bars: 1,000 μm in low-magnification images and 250 μm in high-magnification images. (B) Photomicrographs of brain lesions in vestibular nuclei and reticular nucleus from Sqor^ΔN/ΔN mouse no. 3. Prussian blue staining was used to detect subacute hemorrhage. Blue stain indicates the location of previous episodes of bleeding. An enlargement of the boxed regions is provided below each cross section. Scale bars: 500 μm in low-magnification images and 200 μm in high-magnification images. (C) Photomicrographs of fresh hemorrhages surrounding small vessels in the white matter. Hematoxylin and eosin staining was performed to detect acute hemorrhage. Fresh hemorrhages surrounding small vessels are indicated by black arrows. Scale bars: 50 μm. (D) Brain partial pressure of oxygen was measured using an optical oxygen probe at the reticular nucleus and caudoputamen of Sqor^ΔN/ΔN mice and control mice at postnatal age 40 days. Comparisons were made using unpaired 2-tailed t test. n = 4 mice for each group. Data are presented as means with SD. GFAP, glial fibrillary acidic protein; Iba-1, ionized calcium-binding adaptor molecule 1; NeuN, neuronal nuclei; PO₂, partial pressure of oxygen.

Figure 5. Metronidazole decreased systemic H₂S levels, alleviated Leigh-like disease, and prolonged the lifespan of Sqor^ΔN/ΔN mice.

(A) Kaplan-Meier survival probability curve of Sqor^ΔN/ΔN mice treated with metronidazole or DMSO (4.4%). Log-rank P value and sample size are shown. (B) Body weight trajectory of Sqor^ΔN/ΔN mice treated with metronidazole or DMSO. n = 10 mice for each group. (C) Body temperature trajectory of Sqor^ΔN/ΔN mice treated with metronidazole or DMSO. n = 10 mice for each group. (D) Results of the rotarod test at postnatal ages 20, 30, 40, and 50 days in Sqor^ΔN/ΔN mice treated with metronidazole or DMSO. Two-tailed t test. n = 5–10 mice for each group. (E) Blood lactate levels at postnatal ages 30, 40, and 50 days in Sqor^ΔN/ΔN mice treated with metronidazole or DMSO. Comparisons between Sqor^ΔN/ΔN mice treated with metronidazole or DMSO were made using unpaired 2-tailed t test. Mixed-effects analysis with Dunnett’s multiple-comparison test was performed to compare blood lactate levels at postnatal ages. n = 4–5 mice for each group. (F) Sulfide levels in feces, brain, liver, and muscle in Sqor^ΔN/ΔN mice treated with metronidazole or DMSO and control mice treated with DMSO (the sulfide level for control mice treated with DMSO was set to 1). Data were analyzed using Kruskal-Wallis test with Dunn’s multiple-comparison test for feces, and 1-way ANOVA with Dunnett’s multiple-comparison test for brain, liver, and muscle. n = 6–7 mice for each group. Data are presented as medians with interquartile range for feces. (G) Histochemical staining was used to assess COX activity in a Sqor^ΔN/ΔN mouse treated with metronidazole or DMSO and a control mouse treated with DMSO. Brown color indicates cytochrome c oxidase activity levels. Scale bars: 100 μm. MNZ, metronidazole. Data are presented as means with SD unless indicated otherwise.

Figure 6. A sulfur-restricted diet decreased the systemic H₂S levels, alleviated the clinical manifestations, and prolonged the lifespan of Sqor^ΔN/ΔN mice.

(A) Kaplan-Meier survival probability curve was used to quantify increased survival of Sqor^ΔN/ΔN mice fed a sulfur-restricted diet compared with mice fed a control diet. Log-rank P value and sample size are shown. (B) Body weight trajectory of Sqor^ΔN/ΔN mice fed a sulfur-restricted diet or a control diet. n = 10 mice for each group. Data are presented as means with SD. (C) Body temperature trajectory of Sqor^ΔN/ΔN mice fed a sulfur-restricted diet or a control diet. n = 10 mice for each group. Data are presented as means with SD. (D) A rotarod was used to measure motor function of Sqor^ΔN/ΔN mice fed a sulfur-restricted diet or a control diet at postnatal ages 20, 30, 40, and 50 days. Comparisons were made using unpaired 2-tailed t test. n = 3–10 mice for each group. Data are presented as means with SD. (E) Compared with Sqor^ΔN/ΔN mice fed a control diet, blood lactate levels in Sqor^ΔN/ΔN mice fed a sulfur-restricted diet were lower at postnatal ages 40 and 50 days. Comparisons between Sqor^ΔN/ΔN mice fed a sulfur-restricted diet or a control diet were made using unpaired 2-tailed t test. Mixed-effects analysis with Dunnett’s multiple-comparison test was performed to compare blood lactate levels at postnatal ages 30, 40, and 50 days. n = 4–6 mice for each group. Data are presented as means with SD. (F) Sulfide levels in feces, brain, liver, and muscle were measured using HSip-1. Sulfide levels in Sqor^ΔN/ΔN mice fed a sulfur-restricted diet or a control diet were compared with those in control mice fed a control diet (the sulfide level for control mice fed a control diet was set to 1). Data were analyzed using Kruskal-Wallis test with Dunn’s multiple-comparison test for feces and muscle, and 1-way ANOVA with Dunnett’s multiple-comparison test for brain and liver. n = 5–7 mice for each group. Data are presented as medians with interquartile range for feces and muscle, and as means with SD for brain and liver. CD, control diet; SRD, sulfur-restricted diet.