Three-dimensional DNA

3D genomics helps to understand how our genes work

Sergey Razin, Corresponding Member of the Russian Academy of Sciences, Professor, Head of the Department of Molecular Biology of the Faculty of Biology of Moscow State University, Head of the Laboratory of the Institute of Gene Biology of the Russian Academy of Sciences
Sergey Ulyanov, Candidate of Biological Sciences, Senior Researcher at the Faculty of Biology of Moscow State University and Institute of Gene Biology of the Russian Academy of Sciences
Kommersant Nauka No. 5, July 2017
Published on the website "Elements"

Studies of the spatial configuration of DNA in chromosomes have revealed unexpected, previously unknown causes of severe human diseases.

The emergence of three-dimensional genomics

Over the decades that have passed since the proof of the genetic function of DNA in the forties of the last century, the idea remained unchanged that the measure of the distance between any parts of the genome is the length of the DNA chain separating them. Today we know that the ability of DNA to form loops and other complex structures enables genes and genome elements that control their work (enhancers) to be close to each other in the space of the cell nucleus, even if they are separated by an extended DNA fragment (Fig. 1).

In recent years, new approaches have emerged that make it possible to study the laying of genomic DNA in the cell nucleus. This marked the beginning of the development of the scientific direction, which we call 3D genomics. Using these approaches, it was shown that chromosomes are divided into structural and functional blocks - topologically associated domains (TADs). Sections of the genome from one TAD come into contact with each other much more often than with sections from neighboring TADs. This makes it possible to represent TADs in the form of relatively dense tangles of DNA strands. The results of many experiments show that an enhancer can activate only genes located inside the TAD where its enhancer is located.

Fig. 1. When a DNA loop is formed, even genes located far away in its chain and the genome elements controlling their work turn out to be close to each other

Thus, TADs play an important role in controlling gene activity. The removal or damage of the DNA section separating neighboring TADs leads to the fact that the enhancer gets the opportunity to activate genes that normally do not work in this type of cell, which can cause serious diseases such as cancer, disorders of the formation of sexual characteristics and failures in the development of the embryo (Fig. 2).

Fig. 2. Contacts between enhancers and promoters, as a rule, are formed within one TAD. Removing the boundary between adjacent TADs leads to their merging. In this case, the enhancer gets the opportunity to contact genes whose work is normally not controlled by this enhancer, which is fraught with the development of pathologies, for example, cancer

But what ensures the division of the genome into TADs? The work of our laboratory has made a significant contribution to solving this problem. We found that the organization of genomic DNA in TADs occurs largely spontaneously and is regulated by simple physical laws. Our work was published in the prestigious international journal Genome Research (Sergey V. Ulianov et al. Active chromatin and transcription play a key role in chromosome partitioning into topologically associating domains // Genome Research, 2016, 26, p. 70-84), it has been much talked about in the press and on television.

The essence of our results lies in the fact that the boundaries of the TADs are the sections of the genome containing the "household" genes, that is, genes that work in all types of cells and are necessary to maintain basic cellular processes. Due to a number of features, such sections of the genome cannot collapse into dense globules, thereby creating a "marking" of the boundaries of TADs in the genome.

It is important to note that in addition to various biochemical techniques, we used genome structure modeling on the Lomonosov Moscow State University supercomputer, and the results of this modeling clearly indicate that DNA stacking in individual cells can vary quite a lot (Fig. 3).

Fig. 3. We can investigate the spatial organization of the genome using supercomputers to simulate the structure of virtual polymers, similar in physical properties to the real genome. The figure shows the spontaneous folding of a virtual polymer, guided by the chaotic movement of molecules and electrostatic interactions

From cell populations to individual cells

In the vast majority of cases, hundreds of thousands and even millions of cells have to be used in molecular biological research in each experiment. This is due to the fact that there are very few studied molecules in one cell, and this makes it extremely difficult to work with them.

For example, the amount of genomic DNA in one human cell is about one hundred thousand million times less than one gram. Working with a large number of cells leads to the fact that the results obtained in the experiment, as a rule, allow us to establish the average, most typical values of certain parameters of cellular physiology. In a sense, the information obtained can be likened to the "average temperature" of patients in a hospital.

Of course, the results of studies of cell populations have allowed us to establish many important patterns. However, it is well known that cells of the same type that look exactly the same under a microscope can differ in many different biochemical parameters. Studies of the work of the genome in one single cell are becoming a "trend of the time" and have already made a significant contribution to understanding how the fine-tuning of the work of our genome is carried out. Such studies also influence the development of medicine, since, for example, events occurring in a very small proportion of cells can give rise to the development of tumors. When studying large cell populations, such events often go unnoticed.

In collaboration with Austrian and American colleagues, we have developed a new experimental approach that allows us to analyze the genome stacking in individual cells. Using this approach, we were able to construct significantly more detailed maps of the spatial organization of the mouse genome than in the previous work of our English colleagues. An analysis of the data obtained, recently published in the journal Nature (Ilya M. Flyamer et al. Single-nucleus Hi-C reveals unique chromatin reorganization at oocyte-to-zygote transition // Nature, 544, p. 110-114), provided strong evidence that genome stacking differs significantly in individual cells (Fig. 4). In our opinion, this indicates that that there is a constant search of various genomic configurations in the cell — and this makes it possible to quickly adapt to changing environmental conditions.

Fig. 4. The spatial organization of the same part of the genome can vary significantly in different cells of the same population. So, it can be folded into one dense globule (TAD3+4) or divided into several globules (TAD3 and TAD4)

Although in most cases it is easier to study a cell population than individual cells, for some types of cells, the population approach cannot be used at all, because these cells are, as they say, a piece product. Using the experimental approach developed by us, we were able to study the stacking of the paternal and maternal genomes in fertilized mouse eggs (zygotes).

Quite unexpectedly, we discovered that the laying of genomic DNA in the maternal nucleus in the zygote is fundamentally different from the laying of the genome in the nuclei of any other cell type. In the nuclei of all other studied cell types, active and "silent" regions of the genome are spatially separated from each other. In the maternal nucleus of the zygote, on the contrary, this is not observed. Our results suggest that the configuration of the genome in the maternal nucleus is the most basic, corresponding to the so-called state of totipotence, which allows to obtain many different cell types of an adult organism from one zygote during embryonic development.

The spatial configuration of the genome in the maternal nucleus is the most basic and allows to obtain many different types of cells of an adult organism from one fertilized egg

3D Genomics and Medicine

Discussing the news of molecular biology, as a rule, they talk about the "human genome", or "human genomic DNA", or simply about DNA. But it is important to remember that the nuclei of cells in our body normally contain 23 different DNA molecules, each of which forms a separate chromosome, and together they are called a genome.

Each chromosome is laid in a certain, unique way for it and is located in the nucleus of the cell so that the territory occupied by it practically does not intersect with the territories of neighboring chromosomes. In this sense, the cell nucleus resembles the globe, on which there are many states occupying certain territories and separated by borders.

History knows many examples of how events in one state directly affected life in neighboring countries and world politics in general. In the cell nucleus, the situation is about the same. Any changes in the work of the genome, whether it is the launch or suppression of the expression of individual genes, or the appearance of extra copies of certain chromosomes, can affect the work of genes that are not directly affected by these changes and are located in other chromosome states.

As an example, we can point to the results of the work that we have done with our French colleagues from the Gustave Roussy Institute. The results of this work were published in the prestigious hematology journal Blood (Jeanne Allinne et al. Perinucleolar relocalization and nucleolin as crucial events in the transcriptional activation of key genes in mantle cell lymphoma // Blood, V. 123, 13, p. 2044–2053, doi: 10.1182/blood-2013-06-510511). We have convincingly demonstrated that simply moving a certain gene from one region of the nucleus to another can cause its activation in cells where it normally does not work. This triggers a cascade of processes that eventually leads to the development of leukemia, the root causes of which would be difficult to understand without taking into account the spatial structure of the genome.

It is important to note that the disclosure of a fundamentally new mechanism for the occurrence of leukemia creates the basis for the development of ways to combat these diseases. Thus, the research of genomic DNA stacking in the nucleus is of interest not only for fundamental science, but also for medicine, contributing to a deeper understanding of the mechanisms of various pathologies.

Evolution of 3D genome organization

Since the three-dimensional organization of the genome is one of the tools for regulating gene expression, it should be an object of evolution. In a recent work carried out in our laboratory, the results of which are published in the highly rated international journal Molecular Biology and Evolution (Anastasia P. Kovina et al. Evolution of the Genome 3D Organization: Comparison of Fused and Segregated Globin Gene Clusters // Molecular Biology and Evolution, V. 34, 6, p. 1492–1504, doi: 10.1093/molbev/msx100), we have shown that this is indeed the case.

Using the example of the evolution of clusters of globin genes of vertebrates, we demonstrated that as they move up the evolutionary ladder, linear segments of chromosomes are lost, while segments organized into globules (tangles) are preserved (Fig. 5).

Fig. 5. The results of our studies of the spatial structure of the locus of globin genes of the tropical fish Danio rerio show that the genome segments that facilitate the establishment of contacts between enhancers and genes (organized in loops) are preserved during evolution. Linear segments of the genome, on the contrary, are discarded by natural selection

Most likely, this is due to the fact that in mammals the role of remote enhancers in the regulation of gene activity increases significantly. The establishment of contacts between such enhancers and the genes controlled by them is ensured by the formation of DNA loops, which leads to the formation of globules.

Final notes

In recent years, domestic science has often and in many cases been reasonably criticized for low productivity and lack of international-level work. We have shown above how one relatively small domestic laboratory successfully works at the forefront of world science, systematically publishing the results of its work in the most prestigious international journals.

The implementation of all the above works became possible thanks to a large grant from the Russian Science Foundation. It is difficult to overestimate the importance of such support, not only because it provides the opportunity to perform expensive work, such as massive DNA sequencing. But even more importantly, such grants provide an opportunity to attract young researchers to work, providing a reasonable alternative to going abroad. At least in experimental biology, targeted support of teams working at the world level (as can be judged by the presence of publications in rating international journals) is, in our opinion, the most direct way to the revival of science in our country.

Portal "Eternal youth" http://vechnayamolodost.ru  05.09.2017