Transcriptional Activators and Large-scale Chromatin Decondensation

Reference:

Tumbar, T., Sudlow, G., and Belmont, A.S. (1999) "Large-scale chromatin unfolding and remondeling induced by VPI6 acidic activation domain," J. Cell Biol. 145:1341-54

Tsukamoto, Hashiguchi, N., Janicki, S., Tumbar, T., Belmont, A.S., and Spector, D.L. (2000) "Visualization of gene activity in living cells," Nature Cell Biology 2:871-878.

Background:

The effects of transcriptional activators on local chromatin structure is now well documented. Many transcriptional activators have been shown to have histone modifying activities or to recruit other protein complexes which have histone modifying activities. What is not clear is the effect transcriptional activators might have on higher levels of chromatin organization.

Previously, we described the folding of 30 nm chromatin fibers into still larger scale fibers, in the range of 60-130 nm in diameter, which we have called chromonema fibers (see Large-scale chromatin structure and chromonema fibers:) Given the large molecular weights of many macromolecular protein complexes involved in transcriptional regulation, we wondered what effect this large-scale chromatin organization might have on transcriptional activation.

One possibility was that during the process of transcriptional activation, an activity which decondenses large-scale chromatin organization might be recruited by some transcriptional activators.

Results:

To test this possibility, we designed an experiment to observe the effect on large-scale chromatin organizaton of targeting a large amount of a transcriptional activator to a specific chromosome region. We exploited a system we had developed previously to create gene amplified chromosome regions containing lac operator direct repeats which we could visualize using lac repressor binding (see In vivo tagging of chromosome sites and regions). We chose a cell line that contained an ~ 90 Mbp amplified chromosome region which shared some properties of heterochromatin. Specifically it remains highly condensed through most of interphase and replicates in middle to late S phase.

We targeted the acidic activation domain (AAD) of the strong viral transcription factor, VP16, using a fusion protein of lac repressor, or GFP-lac repressor, with the VP16 AAD. This targeting was accomplished either by microinjecting purified GFP-lac repressor- VP16 AAD protein into cells or by transient transfection.

A dramatic decondensation of the heterochromatic amplified chromosome region was observed. Interestingly, this decondensation occurred through an extension and straightening of chromonema fibers such that the condensed chromosome arm unfolded into an ~ 25 um long chromonema fiber. This unfolding was accompanied by a high level of transcriptional activation. Decondensation was still observed after inhibition of transcription, implying that decondensation was not simply the result of high transcriptional activity. Histone acetylation and recruitment of histone acetyltransferases, including GCN5, CBP/p300, and PCAF was observed.

Decondensation of a heterochromatin amplified chromosome arm is observed after targeting the acidic activation domain of the VP16 viral transcription factor. DNA is stained with DAPI (blue) and the lac repressor_VP16 fusion protein is in red. A cell containing an amplified chromosome region containing lac operator repeats was transiently transfected with a lac repressor- VP16 AAD expression vector (see text above). Decondensation appears to occur through the uncoiling of a ~80 nm chromonema fiber.

Conclusion:

Our results demonstrate that targeting large-amounts of a transcriptional factor can lead to large-scale chromatin decondensation. Although our system is artificial, we propose that it may represent an exaggerated response which occurs on a smaller scale adjacent to natural gene loci. In this model, changes in large-scale chromatin organization might be induced by elements of the transcriptional machinery over distances of tens to hundreds of kb. Experiments are in progress to test this model. Key questions include whether this decondensation occurs through action of the same type of chromatin modifying activities know to operator on local nucleosome structure, or whether novel chromatin modifying activities are involved. Interestingly, high levels of transcription occur although the chromatin remains packaged in large-scale chromatin fibers. This may be characteristic for transcription from many pol 2 genes within diploid cell nuclei, as suggested by in situ localization of transcription adjacent to condensed chromatin domains.