Structure and regulation of DNA condensates by disordered linker histone tails

对控制基因转录活性的基本因素的研究是所有生物体的基本生物学基础，只有很好地描述浓缩基因组DNA(染色质)的物理性质，才能正确地理解这些过程。有鉴于此，或许令人惊讶的是，还有这么多东西需要理解。两米长的DNA被浓缩到每个细胞核中。这是分两个阶段实现的。第一个阶段是将DNA包裹在蛋白质卷轴上，形成类似于细绳上的珠子的“核小体”。第二个阶段涉及将这种结构进一步浓缩成一个更紧凑的结构；这个阶段还没有得到很好的理解。生物学可以自上而下(即观察细胞)或自下而上(即观察原子级分子)；人们总是希望它们结合在一起形成一致的图景，从而克服固有的限制，并验证每种方法产生的模型。这两种方法都是完全理解大多数生物过程所必需的。就染色质而言，目前存在关于第二阶段缩合的不连续，因为应用于活细胞的最好的新成像技术尚未观察到自下而上方法预测的第二阶段形成的30 nm纤维。相反，纤维看起来更加开放、灵活、无序和异质，自组装成具有液体性质的大染色质球体。看起来，高度有序的30 nm纤维可能代表了染色质浓缩和不活跃的极端情况，而且与转录更相关的情况可能要动态得多。这一观点与我们和其他人从不同的自下而上的方法中产生的一些不同的观点是一致的，这些观点拾取了包装DNA的蛋白质中高度固有的无序，以及它们将DNA浓缩成稠密液体冷凝物的能力，在这种凝聚体中，纤维的动力学被保留了下来。液体冷凝物是染色质对环境刺激快速反应的一种引人注目和更可信的方法。我们开发了一种模型系统，使我们能够在原子水平上研究DNA包装蛋白质-特别是“连接子组蛋白”-与DNA结合的方式，以及蛋白质/DNA复合体相分离成致密液滴的条件。它还允许进行热力学测量。我们想利用这个系统来回答几个关键问题：DNA到底是如何被连接子组蛋白结合和缩合的？所产生的缩合物的高度拥挤的环境是否允许已知在体内起作用的修饰酶进入并发挥作用，以及缩合物如何反应？蛋白质/DNA复合体的内在无序如何促进其他染色质相关蛋白质的快速交换？缩合物中存在哪些更高级别的结构，以及它们的组装是如何控制和调节的？许多连接物组蛋白亚型(例如，在人类中有11个细微不同的连接物组蛋白)和DNA序列在内容和修饰上有什么不同和相似之处，例如那些标志着基因开始的序列，或者那些被添加了甲基团的DNA序列。总而言之，基因组的缩合是一个基本的过程，通过各种不同的机制发生在生命的各个王国。总的来说，真核生物实现这一目标经历了两个阶段。第一种是通过核小体的形成，这已经得到了深入的研究和很好的理解。第二个是它通过连接子组蛋白的进一步缩合，这一点还不太清楚，将在本提案中解决。

Study of the fundamental factors controlling the transcriptional activity of genes underpins the basic biology of all organisms, and the processes can only be properly understood if the physical nature of condensed genomic DNA (chromatin) is well described. Given this, it is perhaps surprising that there is so much left to understand. Two meters of DNA are condensed into each cell nucleus. This is achieved in two stages. The first stage involves wrapping the DNA around protein spools to form "nucleosomes" that resemble beads on a string. The second stage involves the further condensation of this structure into one that is more compact; this stage is less well understood.Biology can be approached from the top down (i.e. looking at cells) or bottom up (i.e. looking at atomic-level molecules); the hope is always that they join up to make a consistent picture, thus overcoming the inherent limitations and validating the model generated from each approach. Both approaches are necessary to fully understand most biological processes.In the case of chromatin, there is currently a discontinuity concerning the second stage of condensation, since the 30 nm fibre predicted to form in the second stage by a bottom-up approach has not been observed by the best new imaging techniques applied to live cells. Instead, the fibre appears more open, flexible, disordered and heterogeneous, self-assembling into large chromatin globules with liquid-like properties. It seems likely that the highly ordered 30 nm fibre may represent an extreme case of condensed and inactive chromatin, and the more transcriptionally-relevant situation may be much more dynamic. This view is resonant with some alternative views that are emerging from different bottom-up approaches by us and others, that pick up on the high level of inherent disorder in the proteins that package DNA, and their ability to concentrate the DNA into dense liquid condensates, in which the dynamics of the fibre are retained. A liquid condensate is a compelling and more plausible means by which chromatin could respond quickly to environmental stimuli.We have developed a model system that allows us to study, at atomic-level resolution, the way DNA-packaging proteins - specifically the 'linker histones' - bind to the DNA, and the conditions under which the protein/DNA complexes phase separate into dense liquid droplets. It also permits thermodynamic measurements. We would like to exploit this system to answer several key questions: exactly how is DNA bound and condensed by linker histones?; does the highly crowded environment of the resulting condensate allow the entry and action of modifying enzymes known to act in vivo, and how do the condensates respond?; how does the intrinsic disorder of the protein/DNA complexes facilitate rapid exchange by and with other chromatin-associated proteins?; what higher-order structures are present in the condensates, if any, and how is their assembly controlled and regulated?; what are the mechanistic similarities/differences in genome condensation between the many linker histone subtypes (there are for example 11 subtly different linker histones in humans that either come and go through the lifetime of a cell, or locate to sperm or egg) and DNA sequences differing in content and modifications, such as those marking the start of genes or those that are 'silenced' by the addition of methyl groups?In summary, the condensation of the genome is a fundamental process that occurs by a variety of mechanisms across the kingdoms of life. By and large, eukaryotes achieve it in two stages. The first is through the formation of nucleosomes, which has been heavily studied and is well understood. The second is its further condensation by linker histones, which is less well understood, and will be addressed by this proposal.