Mutation in mitochondrial complex I ND6 subunit is associated with defective response to hypoxia in human glioma cells

Abstract

Background: Hypoxia-tolerant human glioma cells reduce oxygen consumption rate in response to oxygen deficit, a defense mechanism that contributes to survival under moderately hypoxic conditions. In contrast, hypoxia-sensitive cells lack this ability. As it has been previously shown that hypoxia-tolerant (M006x, M006xLo, M059K) and -sensitive (M010b) glioma cells express differences in mitochondrial function, we investigated whether mitochondrial DNA-encoded mutations are associated with differences in the initial response to oxygen deficit.

Results: The mitochondrial genome was sequenced and 23 mtDNA alterations were identified, one of which was an unreported mutation (T-C transition in base pair 14634) in the hypoxia-sensitive cell line, M010b, that resulted in a single amino acid change in the gene encoding the ND6 subunit of NADH:ubiquinone oxidoreductase (Complex I). The T14634C mutation did not abrogate ND6 protein expression, however, M010b cells were more resistant to rotenone, an agent used to screen for Complex I mutations, and adriamycin, an agent activated by redox cycling. The specific function of mtDNA-encoded, membrane-embedded Complex I ND subunits is not known at present. Current models suggest that the transmembrane arm of Complex I may serve as a conformationally driven proton channel. As cellular respiration is regulated, in part, by proton flux, we used homology-based modeling and computational molecular biology to predict the 3D structure of the wild type and mutated ND6 proteins. These models predict that the T14634C mutation alters the structure and orientation of the trans-membrane helices of the ND6 protein.

Conclusion: Complex I ND subunits are mutational hot spots in tumor mtDNA. Genetic changes that alter Complex I structure and function may alter a cell's ability to respond to oxygen deficit and consolidate hypoxia rescue mechanisms, and may contribute to resistance to chemotherapeutic agents that require redox cycling for activation.

Figure 1

mtDNA copy number and proteins in M010b cells. A, mtDNA copy number in GBM cell lines (M010b, M059K, M006x, M006xLo) and a normal fibroblast cell strain (GM38) by quantitative PCR. Graphical analyses of cytochrome b (■) and β-actin (▼) amplified from serially diluted template DNA. Standard curves were derived from three or more independent PCR experiments. Error bars show mean ± s.d. B, ³⁵S-labeled mitochondrial translation products in M010b cells. The ND6 protein (indicated by the arrow) was identified on the basis of its relative position within the electrophoretic migration pattern of the 13 mtDNA-encoded proteins [25].

Figure 2

Complex I Activity. A flow cytometric assay of overall Complex I activity in normal human fibroblasts (GM38) and hypoxia-tolerant (M006x, M006xLo, M059K) and -sensitive (M010b) GBM cells. Changes in the relative ratio of dihydroethidium fluorescence after treatment of cells with rotenone are shown. Data shown for five or more separate experiments for each cell line. Error bars show mean ± s.d.

Figure 3

M010b growth assay. Cell growth assay for M010b cells grown for 1–8 days in DMEM/medium containing either 10 mM glucose (▲) or 5 mM galactose (■) as substrates. Data are from three independent experiments. Error bars show mean ± s.d.

Figure 4

Adriamycin and TTFA cytotoxicity. Effects of adriamycin or TTFA on GBM cell survival. Serial dilutions of A., adriamycin, or B, TTFA were added to GBM cells growing in multi-well plates. Cell viability was determined 96 h later using a MTT assay. Error bars show mean ± s.d. from three or more independent experiments. When not visible, errors were smaller than the symbol. ▲, M010b; ■, M006x; ●; M006xLo; ◆, M059K.

Figure 5

HIF-1α protein expression. Western blot analysis of HIF-1α protein expression in hypoxia-tolerant (M006x) or hypoxia-sensitive (M010b) cells. Lanes 1,4, cells incubated (24 h) then lysed under ambient aerobic conditions (18% O₂); Lanes 2,5, cells incubated in hypoxia (0.6% O₂ × 24 h) followed by 1 h re-oxygenation prior to lysis; Lanes 3,6, cells incubated in hypoxia (0.6% O₂ × 24 h) then lysed in an atmosphere of 1% O₂.

Figure 6

ND6 protein secondary structure. Predicted secondary structure for human ND6 protein. A, Hydropathy plot for the human ND6 protein. Red line: transmembrane helix preference; Blue line: beta-turn preference; Gray line: modified hydrophobic moment index; Dark red line: predicted transmembrane helix. B, The secondary structures predicted for the complete ND6 proteins. The models predict six hydrophobic domains (putative hydrophobic transmembrane helices) that are connected to each other by five loops (two external loops and three internal loops). The secondary structure of the normal protein is shown for reference; changes in secondary structure predicted to occur as a result of the mutation M14V in M010b cells; the M64V mutation in LHON and the A72V mutation in LHON are illustrated.

Figure 7

Homology-based modeling of ND6 proteins. Homology-based modeling of normal and mutated ND6 3D protein structures. By comparison to the predicted 3D structure of the normal ND6 protein, these models predict that the M14V mutation in M010b cells, the M64V mutation in LHON, and the A72V mutation in LHON would significantly disrupt the orientation of the entire ND6 protein within the mitochondrial membrane, and would alter the interactions of the individual helices of each protein.