Centrosomes isolated from various sources, including human cells, have the capacity to induce parthenogenetic development when injected into unfertilized amphibian eggs. We recently isolated calf thymus centrosomes and showed that they differ structurally and functionally from previously isolated centrosomes of KE37 cells, in that the two centrioles in calf thymocytes are linearly associated by their proximal ends through a mass of electron dense material and nucleate few microtubules from their distal ends (Komesli, S., F. Tournier, M. Paintrand, R. Margolis, D. Job, and M. Bornens. 1989. J. Cell Biol. 109:2869-2878). We report here that these centrosomes are also unable to induce egg cleavage and examine the various possibilities which could account for this lack of competence. The results show that: (a) the kinetics of microtubule assembly on calf thymus centrosomes in Xenopus extracts are comparable to those of KE37 centrosomes; (b) centrosomes isolated from thymus of calves raised under controlled conditions (without anabolic agents) also lack competence; (c) centrosomes isolated from bovine cells of other tissues are competent; (d) centrosomes isolated from thymus of three other species (rat, mouse, and human) are competent. Since the lack of activity of calf thymus centrosomes apparently was not linked to species or tissue differences, we compared the ultrastructure of the centrosomes in the various centrosome preparations. The results show a strict correlation between the linear arrangement of centrioles and the lack of activity of the centrosomes. They suggest that the centrosome cycle can be blocked when the centrioles are prevented from separating into a nonlinear configuration, a step which might be critical for the initiation of procentriole budding. They also indicate that the centrosome may be involved in the G0-G1 transition.

Centrosomes from calf thymocytes were isolated using a simple preparative procedure that provides large yields of free organelles. A comparative study with centrosomes isolated from human cultured lymphoblasts has led to the discovery of important differences in the structure of the two isolates and in their capacity to nucleate microtubules from purified tubulin. The possibility that the centrosomal structure depends upon the growth state of cells is discussed.