Mechanisms that regulate the number of cells constituting the body have remained largely elusive. We approached this issue in the ascidian, Halocynthia roretzi, which develops into a tadpole larva with a small number of cells. The embryonic cells divide 11 times on average from fertilization to hatching. The number of cell division rounds varies among tissue types. For example, notochord cells divide 9 times and give rise to large postmitotic cells in the tadpole. The number of cell division rounds in partial embryos derived from tissue-precursor blastomeres isolated at the 64-cell stage also varied between tissues and coincided with their counterparts in the intact whole embryos to some extent, suggesting tissue-autonomous regulation of cell division. Manipulation of cell fates in notochord, nerve cord, muscle, and mesenchyme lineage cells by inhibition or ectopic activation of the inductive FGF signal changed the number of cell divisions according to the altered fate. Knockdown and missexpression of Brachyury (Bra), an FGF-induced notochord-specific key transcription factor for notochord differentiation, indicated that Bra is also responsible for regulation of the number of cell division rounds, suggesting that Bra activates a putative mechanism to halt cell division at a specific stage. The outcome of precocious expression of Bra suggests that the mechanism involves a putative developmental clock that is likely shared in blastomeres other than those of notochord and functions to terminate cell division at three rounds after the 64-cell stage. Precocious expression of Bra has no effect on progression of the developmental clock itself.

Mechanisms that regulate the number of cell division rounds during embryogenesis have remained largely elusive. To investigate this issue, we used the ascidian, which develops into a tadpole larva with a small number of cells. The embryonic cells divide 11.45 times on average from fertilization to hatching. The number of cell division rounds varies depending on embryonic lineages. Notochord and muscle consist of large postmitotic cells and stop dividing early in developing embryos. Here we show that conversion of mesenchyme to muscle cell fates by inhibition of inductive FGF signaling or mis-expression of a muscle-specific key transcription factor for muscle differentiation, Tbx6, changed the number of cell divisions in accordance with the altered fate. Tbx6 likely activates a putative mechanism to halt cell division at a specific stage. However, precocious expression of Tbx6 has no effect on progression of the developmental clock itself. Zygotic expression of a cyclin-dependent kinase inhibitor, CKI-b, is initiated in muscle and then in notochord precursors. CKI-b is possibly downstream of tissue-specific key transcription factors of notochord and muscle. In the two distinct muscle lineages, postmitotic muscle cells are generated after 9 and 8 rounds of cell division depending on lineage, but the final cell divisions occur at a similar developmental stage. CKI-b gene expression starts simultaneously in both muscle lineages at the 110-cell stage, suggesting that CKI-b protein accumulation halts cell division at a similar stage. The difference in the number of cell divisions would be due to the cumulative difference in cell cycle length. These results suggest that muscle cells do not count the number of cell division rounds, and that accumulation of CKI-b protein triggered by tissue-specific key transcription factors after cell fate determination might act as a kind of timer that measures elapsed time before cell division termination.

During embryonic induction, the responding cells invoke an induced developmental program, whereas in the absence of an inducing signal, they assume a default uninduced cell fate. Suppression of the default fate during the inductive event is crucial for choice of the binary cell fate. In contrast to the mechanisms that promote an induced cell fate, those that suppress the default fate have been overlooked. Upon induction, intracellular signal transduction results in activation of genes encoding key transcription factors for induced tissue differentiation. It is elusive whether an induced key transcription factor has dual functions involving suppression of the default fates and promotion of the induced fate, or whether suppression of the default fate is independently regulated by other factors that are also downstream of the signaling cascade. We show that during ascidian embryonic induction, default fates were suppressed by multifold redundant mechanisms. The key transcription factor, Twist-related.a, which is required for mesenchyme differentiation, and another independent transcription factor, Lhx3, which is dispensable for mesenchyme differentiation, sequentially and redundantly suppress the default muscle fate in induced mesenchyme cells. Similarly in notochord induction, Brachyury, which is required for notochord differentiation, and other factors, Lhx3 and Mnx, are likely to suppress the default nerve cord fate redundantly. Lhx3 commonly suppresses the default fates in two kinds of induction. Mis-activation of the autonomously executed default program in induced cells is detrimental to choice of the binary cell fate. Multifold redundant mechanisms would be required for suppression of the default fate to be secure.