|
内容記述 |
The body of vertebrates is composed of many repeated structures such as vertebrae,<br />ribs, skeletal muscles and subcutaneous tissues. These are based on transient<br />metameric structures, called somites, which are produced sequentially at certain<br />spatiotemporal intervals in an anterior to posterior sequence concomitant with the<br />posterior elongation of presomitic mesoderm (PSM). The periodicity is regulated by the<br />segmentation CLOCK which undergoes the oscillation of <i>Hes</i> gene expression under<br />the control of Notch signaling within a cell. The traveling wave of the CLOCK is<br />observed from the tail bud region and stops in the anterior PSM where a new somite is<br />generated. This waved pattern is generated by the change of gene expression in each<br />cell and not by cell movement. If the CLOCK oscillates among individual cells without<br />synchrony, the waved pattern will not be generated. Thus, it is required for the additional<br />mechanism which works in a non-cell-autonomous manner to synchronize the CLOCK<br />phase among neighboring cells.<br /><br /> In zebrafish, the CLOCK and its synchronization mechanism has been well<br />understood. A coupled oscillator model was proposed to link these two phenomena; it<br />has been explained that 1) INPUT receives OUTPUT from neighboring cells, 2)<br />Effectors of the CLOCK are activated by the INPUT and operate as CLOCK<br />components, 3) OUTPUT transmits information reflecting its own CLOCK phase to<br />neighboring cells. Therefore, they correct their CLOCKs each other by coupling their<br />CLOCKs. In the zebrafish somitogenesis, it was demonstrated that INPUT is Notch<br />signaling from Notch1a, the CLOCK is Her1/7 that shows oscillation via the<br />negative-feedback mechanism and OUTPUT is DeltaC oscillation controlled by Her1/7.<br /><br /> However, it has been difficult to reveal the synchronization mechanism in mouse<br />somitogenesis because the CLOCK itself disappears in a simple gene knockout mouse<br />that lacks function of Notch signaling, since Notch signal is a core component of the<br />CLOCK. Furthermore, the segmentation CLOCK components involved in the regulation<br />are more complicated in mice as compared with zebrafish. In the mouse PSM, Lfng, a<br />glycosyltransferase that is not implicated in zebrafish, oscillates upon activation by<br />Notch activity and repression by Hes7, and acts as a negative regulator for Notch<br />signaling via modifying Notch1 receptor. Hence, the cyclic expression of Lfng makes a<br />Notch signal oscillation as a segmentation CLOCK.<br /><br /> To reveal the mechanism to generate synchronized CLOCK oscillation in mice, I first<br />examined Dll1 expression pattern that works as an OUTPUT in zebrafish. The results<br />that Dll1 transcripts slightly oscillated in the PSM but its protein did not show clear<br />oscillation indicate that the OUTPUT mechanism of the coupling in mice is different from<br />that of zebrafish. Next, I performed mosaic embryo analyses to clarify the<br />synchronization mechanism. The mosaic analysis using wild-type and gene-knockout<br />(KO) cell is a powerful method to ask the mechanism involved in the cell-cell<br />communication. I conducted two types of mosaic embryo analyses using Dll1-null and<br />Lfng-null cells. If the coupled oscillator model via Notch signaling is utilized to generate<br />the synchronized CLOCK oscillation in mouse somitogenesis, it is thought that INPUT is<br />Notch signaling through Notch1 receptor, the CLOCK is the oscillation of Hes7 and<br />OUTPUT is an unknown factor through a transmitter Dll1. In Dll1-null mosaic embryos,<br />Dll1-KO cells do not have Dll1 which acts as a transmitter of the CLOCK to transmit its<br />own CLOCK state to neighboring cells but they have Notch1 (receiver). Therefore, I<br />expected that Dll1-KO cells must show incomplete coupling with neighboring wild-type<br />cell, but should not interfere synchronized oscillation of the CLOCK among wild-type<br />cells because Dll1-KO cells cannot transmit signals. I found that the CLOCK showed<br />abnormal pattern in Dll1-null mosaic embryos, however, it exhibited synchronized<br />oscillation to some degree. Therefore, the reduction of coupling cells may have caused<br />abnormal CLOCK pattern. On the other hand, CLOCK synchronization will be disrupted<br />in the Lfng-null mosaic embryo if Notch signal regulates synchronized CLOCK<br />oscillation through a coupling mechanism as zebrafish and if Lfng is involved in the<br />coupling mechanism. Lfng-null mosaic embryos showed severer defect in the<br />synchronized CLOCK oscillation compared with the Dll1-null mosaic embryos. These<br />results suggest that Notch signal also exhibits dual roles in the CLOCK and its<br />synchronization through the coupling mechanism as in the case of zebrafish.<br />Surprisingly, Lfng KO cells in Lfng-null mosaic embryos showed either positive or<br />negative Notch activity. This result was unexpected since Notch activity should be<br />up-regulated in the absence of Lfng as expected from the analysis of Lfng KO embryo.<br />Therefore, the oscillation of Notch activity in Lfng KO cells in Lfng-null mosaic embryos<br />must be caused by the presence of wild-type cells that have functional Lfng. These<br />results suggest that Lfng works on Notch signaling via not only <i>cis-</i> but also <i>trans-</i><br />regulation mechanisms and Dll1 activity might be regulated by Lfng. Accordingly, I<br />explored in detail the role of Lfng in the Notch signaling by co-culture experiments using<br />Notch signal reporter luciferase assay. The results indicate that Lfng alter the Notch<br />signaling activity by modifying Dll1 and Notch1.<br /><br /> In this study, I propose a new coupling mechanism to generate synchronized<br />oscillation of segmentation CLOCK in the mouse. It is possible to consider that Lfng can<br />work as the OUTPUT which retains/reflects CLOCK phase information and alters Notch<br />signaling to synchronize CLOCK phase among neighboring cells through the coupling<br />mechanism. Therefore, in mouse somitogenesis, the following five elements are<br />required for the coupling mechanism, 1) INPUT; Notch signaling, 2) CLOCK; the<br />oscillation of Hes7 expression, 3) OUTPUT; Lfng expression reflecting CLOCK phase<br />information, 4) transmitter; Dll1 and 5) receiver; Notchl. In mice, expressions of both<br />Dll1 and Notch1 are not regulated by the CLOCK. |
要旨・審査要旨 (337.3 kB)
