[en] Defects in complex I (NADH:ubiquinone oxidoreductase) are the most frequent cause of human respiratory disorders. The pathogenicity of a given human mitochondrial mutation can be difficult to demonstrate because the mitochondrial genome harbors large numbers of polymorphic base changes that have no pathogenic significance. In addition, mitochondrial mutations are usually found in the heteroplasmic state, which could hide the biochemical effect of the mutation. We propose that the unicellular green alga Chlamydomonas could be used to study such mutations because (1) respiratory-deficient mutants are viable and mitochondrial mutations are found in the homoplasmic state, (2) transformation of the mitochondrial genome is feasible, (3) Chlamydomonas complex I is close to that of humans. To illustrate that, we have introduced a Leu157Pro substitution in the Chlamydomonas ND4 subunit of complex I of two different recipient strains by biolistic transformation, demonstrating that site-directed mutagenesis of the Chlamydomonas mitochondrial genome is possible. This substitution did not lead to any respiratory enzyme defect when it is present in the heteroplasmic state in a patient presenting chronic progressive external ophthalmoplegia. When present in the homoplasmic state in the alga, the mutation does not prevent the assembly of the 950 kDa whole complex I which conserves nearly all the NADH dehydrogenase activity of the peripheral arm. However, the NADH:duroquinone oxidoreductase activity is strongly reduced, suggesting that the substitution could affect ubiquinone fixation to the membrane domain. The in vitro defects are correlated in vivo with a decrease in dark respiration and growth rate.