Mode of Mitochondrial Formation in Hela Cells

The mode of mitochondrial formation has been so far directly investigated only in Neurospora crassa. By using a choline-requMng mutant of this organism, Luck (1963, 1965) has obtained evidence which strongly suggests that mitochondria are formed by growth and division of preexisting mitochondria. In Tetrahymena pyriformis (Parsons and Rustad, 1968), the distribution of long-term [aH]thymidine-labeled mitochondrial DNA among mitochondria during growth in unlabeled medium, as studied by autoradiography, has been found to follow a pattern compatible with this mode of mitochondrial formation. In animal cells, no evidence presently exists regarding the mode of mitochondrial formation, apart from electron micrographs of structures which have been interpreted as mitochondrial fission figures (e.g. Tandler et al., 1969; Larsen, 1970). Both in animal cells (King et al., 1972; Storrie and Attardi, 1973) and in yeast (Linnane and Haslam, 1970) the gross formation of mitochondria has been shown to be independent of mitochondrial protein synthesis. By contrast, the formation of an active cytochrome c oxidase, an inner mitochondrial membrane enzyme, has been found to require the participation of the mitochondrial protein-synthesizing system in HeLa cells (Firkin and Linnane, 1968; Storrie and Attardi, 1972), as in other animal cells and in lower eucaryotic cells (for reviews see Borst, 1972; Sager, 1972). The latter observations have suggested an experimental approach to study of the mode of mitochondrial formation in HeLa cells. Since the number of mitochondria per cell, as recognized by morphological criteria, remains constant for up to four generations in cells growing in the presence of 40 ~g/ml chloramphenicol (a concentration sufficient to block formation of new active enzyme without any effect on the preexisting cytochrome oxidase activity [Storrie and Attardi, 1972]), a constancy in the number of cytochrome oxidase-positive mitochondrial profiles per cell section under these conditions would be in agreement with a mode of mitochondrial formation by growth and division of preexisting mitochondria. Furthermore, the distribution of preexisting cytochrome oxidase activity along the inner mitochondrial membrane, as new mitochondria are formed, should give information concerning the mode of membrane biosynthesis. In the present work, no significant dilution in the number of cytochrome oxidase-positive organelle profiles occurred in the absence of mitochondrial protein synthesis for as many as four cell generations, while the intensity of cytochrome c oxidase activity along the inner mitochondrial membrane decreased uniformly. These results strongly suggest that mammalian cell mitochondria replicate by a random insertion of components into the membranes and subsequent division.