|
内容記述 |
Nucleosome is a fundamental unit of chromatin, consisting of 146bp of DNA wrapped around an octomer of four kinds of histones, H2A, H2B, H3 and H4. Because nucleosomes can be significant obstacles to transcription that are mediated by RNA polymerase II)(Pol II), it is needed to destabilize nucleosomal structure by displacement of histones during Pol II)passage. <br />The destabilization of nucleosomes during transcription results in histone loss over the regions of heavily transcribed genes (Lee, C.K. et al. 2004). Histone loss sometimes causes aberrant transcription initiation from cryptic sites within coding regions (Kaplan, C.D. et al. 2003). To maintain the fidelity of transcription initiation, histone deposition behind Pol II passage is necessary. <br /> Recent studies have revealed an interesting phenomenon regarding histone deposition during transcription in eukaryotes other than yeast. The phenomenon is that histone H3 variant, H3.3 is selectively incorporated into nucleosomes during transcription (Ahmad, K. and Henikoff, S. 2002a). In contrast to canonical histone H3, H3.3 is synthesized throughout the cell cycle and deposited onto DNA both during outside of S phase (Ahmad, K. and Henikoff, S. 2002b, Tagami, H. et al. 2004). The selective deposition of H3.3 is a very interesting fenomenon, however, the underlying mechanisms as well as in vivo biological meanings have been elusive. <br /> To elucidate the mechanisms I tried to identify a factor that mediates the deposition of H3.3. In this study I focused on a transcription elongation factor Spt6 as a candidate, Studies of yeast Spt6 have revealed that Spt6 plays a critical role in maintaining normal chromatin structure during transcription elongation interacting with histone H3 (Bortvin, A. and Winstone, F. 1996, Kaplan, C.D. 2003). Although yeast has no transcription variant of histone H3, these findings in yeast studies led me to an idea that Spt6 might be involved in the variant histone H3.3 deposition during transcriptjon in higher eukaryotes. <br /> In order to understand the Spt6 function in higher eukaryotes, l investigated the in vivo role of Spt6 using <i>D. metanogaster</i> as a model organism. First, to investigate the localizatien of Spt6 on chromatin, I carried out ChIP-on-chip microarray analyses. By the analyses of ChIP-on-chip microarray, strong signals of Spt6 (probe set p-value< 0.001) were dtected on about 30 genes within the half of the entire <i>Drosophila genome</i> ( Table 1 ). As these 30 genes are various in their functions, expression patterns and genome structures, any common features were not found among them. In most of the 30 genes, Spt6 was distributed throughout the each gene, especially in exons ( Figure 3 ). I selected several genes such as <i>heat shock protein 83 (hsp83), La related protein (larp)</i> and <i>thread</i> from these Spt6-localized genes for further analyses. <br /> To investigate the function of Spt6 in vivo I generated and characterized a <i>spt6</i> null mutant.By P-element excision, I generated <i>spt6W40. spt6W40</i> allele has a small deletion in the second<br />exon of the <i>spt6</i> gene that causes a stop of the protein synthesis, hence is functionally null (Figure 5-A). Homozygotes of <i>spt6W40</i> show an embryonic lethal phenotype. <br /> To examine the effects of the <i>spt6</i> mutation on transcription, I performed RT-PCR analyses using RNA from embryos to measure the amounts of mRNA. RT-PCR analyses with primers for 5' sides and 3' sides of <i>hsp83</i> and <i>larp</i> mRNA showed that the levels of the mRNA were increased only at the 3'sides but not the 5' sides in the <i>spt6</i> mutant compared with the wild type(Figure 6-B). These results suggest that the transcription initiates from the middle of the coding region in the <i>spt6</i> mutant. To analyze the size of the transcripts from <i>hsp83</i>, I carried out a Northern blot analysis. The RNA probes synthesized over the coding sequence of <i>hsp83</i> detected not only expected 3 kbp transcripts but also shorter transcripts only in the <i>spt6</i>mutant(Figure 7 ). These shorter transcripts may be the products of aberrant transcription on <i>hsp83</i> gene in the <i>spt6</i> mutant. These results suggest that <i>spt6</i> mutation causes aberrant transcription initiation from cryptic sites within the coding region. <br /> To examine the possibility that the aberrant transcription in the <i>spt6</i> mutant is due to changes in chromatin structure, I carried out a micrococcal nuclease (MNase) assay using <i>spt6W40</i> embryos. Although there was no difference in the patterns of bulk chromatin between the wild type and the <i>spt6 </i>mutant, chromatin on <i>hsp83</i> exhibited an increased sensitivity to MNase in the <i>spt6</i> mutant compared with the wild type(Figure 8). This result indicates that the <i>spt6</i> mutation causes changes in the chromatin structure on the <i>hsp83</i> gene in vivo. Because the ChlP analyses with pan-H3 antibodies showed that the level of general H3 on <i>hsp83</i> was reduced in the <i>spt6</i> mutant(Figure 9), the changes in the chromatin structure on <i>hsp83</i> in the<i>spt6</i> mutant may be due to loss of nucleosomes. <br /> Previously, examination of several <i>Drosophila</i> genes revealed that nucleosomes containing either canonical histone H3 or variant histone H3.3 were lost during transcription, and were selectively replaced with nucleosomes containing H3.3 (Wirbelauer, C. et al. 2005). <br />Because the chromatin structure was changed by the loss of nucleosomes in the <i>spt6</i> mutant, I consider the possibility that Spt6 can help the deposition of H3.3 during transcription. To observe the correlation of Spt6 with H3.3, I measured the levels of H3.3 on three Spt6-enriched genes (<i>hsp83, thread and larp</i>) and two non-enriched genes (β-<i>tubulin</i> and<br /><i>cyp4d21</i>) by the ChIP analyses, The levels of H3.3 were higher on the Spt6-enriched genes rather than non-enriched genes(Figure 10). This result suggests that there is some correlation between Spt6 and H3.3 in their localization. For further examination of physical relationship between Spt6 and H3.3, I carried out a Re-ChIP assay to examine whether Spt6 localizes on nucleosomes containing H3.3 or H3. As the results, on <i>larp</i> and <i>hsp83</i>, Spt6 was efficiently immunoprecipitated with the nucleosomes containing H3.3 rather than H3(Figure11A). This indicates that Spt6 preferentially localizes on the nucleosomes containing H3.3 rather than H3.<br /> In order to examine the functional relationship between Spt6 and H3.3, I used <i>spt6</i> RNAi lines gifted by R.Ueda. When <i>spt6</i> RNAi was ubiquitously induced by an <i>Ay-GAL4</i> driver, almost all the animals died before the pupal stage(Table 2-B). If Spt6 is involved in the deposition of H3.3, a decrease in the efficiency of H3.3 deposition caused by <i>spt6</i> knockdown can be suppressed by over-expression of H3.3. To test this, H3.3 was over-expressed in the <i>spt6</i> RNAi line. As controls H3 or GFP was expressed instead of H3.3.The lethal phenotype of <i>spt6</i> knockdown was significantly suppressed only by the over-expression of H3.3. This result suggests that Spt6 is functionally related with H3.3 in vivo. <br /> In summary I have found that Spt6 plays an important role in maintaining the chromatin structure during transcription elongation, thereby repressing production of aberrant transcripts in <i>D. melanogaster</i>. I also observed the correlation of Spt6 with variant histone H3.3. This is the first report to suggest the relationship between Spt6 and H3.3. Further investigation will clarify the function of Spt6 in the H3.3 deposition, and it may provide new insights into chromatin regulations during transcription elongation. |
要旨・審査要旨 (319.3 kB)
