Infantile encephalopathy and defective mitochondrial DNA translation in patients with mutations of mitochondrial elongation factors EFG1 and EFTu

Abstract

Mitochondrial protein translation is a complex process performed within mitochondria by an apparatus composed of mitochondrial DNA (mtDNA)-encoded RNAs and nuclear DNA-encoded proteins. Although the latter by far outnumber the former, the vast majority of mitochondrial translation defects in humans have been associated with mutations in RNA-encoding mtDNA genes, whereas mutations in protein-encoding nuclear genes have been identified in a handful of cases. Genetic investigation involving patients with defective mitochondrial translation led us to the discovery of novel mutations in the mitochondrial elongation factor G1 (EFG1) in one affected baby and, for the first time, in the mitochondrial elongation factor Tu (EFTu) in another one. Both patients were affected by severe lactic acidosis and rapidly progressive, fatal encephalopathy. The EFG1-mutant patient had early-onset Leigh syndrome, whereas the EFTu-mutant patient had severe infantile macrocystic leukodystrophy with micropolygyria. Structural modeling enabled us to make predictions about the effects of the mutations at the molecular level. Yeast and mammalian cell systems proved the pathogenic role of the mutant alleles by functional complementation in vivo. Nuclear-gene abnormalities causing mitochondrial translation defects represent a new, potentially broad field of mitochondrial medicine. Investigation of these defects is important to expand the molecular characterization of mitochondrial disorders and also may contribute to the elucidation of the complex control mechanisms, which regulate this fundamental pathway of mtDNA homeostasis.

Figure 1.

Brain MRIs of patient I.V. (A and B) and patient S.S. (C and D). A, T2-weighted coronal section showing hyperintense symmetric lesions of the putamen and globus pallidus (arrow). B, T1-weighted image of the same section as in panel A, showing hypointense abnormal signals (arrow) corresponding to those shown in A. C, T2-weighted coronal section showing diffuse hyperintensity of the centra semiovalia and basal ganglia. The arrow indicates an area of micropolygyria in the right perisylvian region. D, T2-weighted coronal section showing diffuse hyperintensity and multiple cysts of the white matter.

Figure 2.

mtDNA-specific protein synthesis on fibroblasts from a control, patient S.S., and patient I.V. The autoradiographic bands are labeled according to standard nomenclature: ND1, ND2, ND3, ND4, ND4L, ND5, and ND6 are subunits of cI; Cyt b is cytochrome b, a subunit of cIII; COI, COII, and COIII are subunits of cIV; and A6 and A8 are subunits of cV.

Figure 3.

Light-microscopy examination of a muscle biopsy specimen from patient I.V. A, Modified Gomori trichrome staining. B, Succinate dehydrogenase (SDH) reaction. C, Cytochrome c oxidase (COX) reaction. D, SDH and COX reaction.

Figure 4.

Molecular characterization of EFG1 and EFTu mutations. Below the sequencing profiles of EFG1 and EFTu genes in patients I.V. and S.S. are the ClustalW interspecies alignments of protein sequences from H. sapiens (Human), Mus musculus (Mouse), Drosophila melanogaster, Caenorhabditis elegans, and S. cerevisiae (Yeast). The mutant nucleotides are in yellow-shaded circles, and the mutant amino acid residues are labeled in red. Western blotting shows the presence of both EFTu and EFTs proteins in patient S.S. The 70-kDa succinate dehydrogenase (SDH) band was immunovisualized as a control.

Figure 5.

Modeling of mitochondrial EFG1 (A and B) and EFTu (C and D). In panel A, the EFG1 GDP-binding site and the wild-type M496 residue are indicated. In panel B, the pocket containing the M496 of EFG1 is magnified, and acidic, basic, and apolar residues are in red, blue, and gray, respectively. In panel C, the three domains and the Arg 339 residue of EFTu are labeled. In panel D, the model structure of human EFTu/tRNA complex is shown. The complex was obtained by superposing the bovine EFTu model on the crystal structure of the EFTu/tRNA complex of T. aquaticus (PDB 1TTT). The tRNA structure is in blue. The binding site for tRNA is magnified; note the position of the R339 residue, which is labeled for clarity.

Figure 6.

A, Tetrads analysis of MEF1/Δmef1 diploid transformed with MEF1^wt or mef1^M516R alleles. The relevant genotype of the diploid strains is indicated on the left. One tetrad for the wild-type MEF1/Δmef1//MEF1^wt transformant and three tetrads for the mutant MEF1/Δmef1//mef1^M516R transformant are shown. Spore clones from the same tetrads are displayed horizontally (a–d). Since deletion of the MEF1 gene was created by insertion of the kanMX4 gene, which confers resistance to G418, the Δmef1 genotype is deduced by the ability to grow on yeast peptone dextrose medium (YPD) supplemented with G418. Since the genetic background is Δura3 and the plasmid pFL38 used for all constructs carries the URA3 wild-type allele, its presence is deduced by the ability to grow in the absence of uracil. A respiratory-proficient phenotype is deduced by the ability to grow on glycerol (Gly) and ethanol (EtOH). Black squares indicate the Δmef1 haploid strains transformed with the wild-type MEF1^wt gene. Black circles indicate the Δmef1 haploid strains transformed with the mef1^M516R mutant allele. B, Tetrads analysis of TUF1/Δtuf1 diploid transformed with TUF1^wt or tuf1^R328Q alleles. The relevant genotype of the diploid strains is indicated on the left. The same experimental scheme described above for MEF1 was applied to TUF1. Black squares indicate the Δtuf1 haploid strains transformed with the wild-type TUF1^wt gene. Black circles indicate the Δtuf1 haploid strains transformed with the tuf1^R328Q mutant allele. C, Oxidative growth phenotype in TUF1 strains. Relevant genotype of the haploid Δtuf1 strains is indicated on the left. Rows 2b, 2d, and 3d refer to the corresponding spores reported in panel B. Equal amounts of serial dilutions of cells from exponentially grown cultures (10⁵, 10⁴, 10³, and 10² cells) were spotted onto YP plates supplemented with 3% Gly or 2% EtOH. The growth was scored after 2 d of incubation at 28°C.

Figure 7.

Reduced versus oxidized cytochrome spectra of MEF1 (A) and TUF1 (B) strains. The peak at 550 nm refers to cytochrome c, the peak at 560 nm refers to cytochrome b, and the peak at 602 nm refers to cytochrome aa3. The cytochrome amount is proportional to the height of the corresponding peak, relative to the baseline of each spectrum.

Figure 8.

Complementation analysis in fibroblasts to determine histochemical and biochemical activities. A, COX-specific cytochemical reaction in immortalized fibroblasts cells from patients I.V. and S.S. and in the same cells after transfection with vectors expressing EFG1^wt and EFTu^wt, grown in DMEM glucose medium, or after exposure for 1 d to DMEM galactose medium. B, Biochemical activities of cI and cIV in immortalized fibroblasts from patients I.V. and S.S. An asterisk (*) indicates Student’s t test P<.05; a double asterisk (**) indicates P<.01. Statistical analysis was performed when four or more values were available for each category.

Figure 9.

Complementation analysis in fibroblasts to determine mitochondrial protein synthesis. pCDNA3 indicates fibroblasts transfected with the pCDNA3 “empty” vector (no insert), EFG1^wt and EFTu^wt indicate fibroblasts transfected with vectors expressing the corresponding wild-type proteins, and Gal refers to fibroblasts exposed to galactose-containing medium. The bands are labeled as in figure 2.

Figure 10.

Complementation analysis in cell hybrids and cybrids. A, COX-specific cytochemical reaction in immortalized fibroblast cells from patients I.V. and S.S., 143B ρ° cells, 143B progenitor cells, transmitochondrial cybrids, and heterodikaryon hybrids. B, Biochemical activities of cIV in fibroblasts, 143B ρ° cells, 143B progenitor cells, transmitochondrial cybrids, and heterodikaryon hybrids.