I can describe to you an interaction that I’ve had in the course of my life—for example, the night that I met a woman playing darts in a New Jersey bar or the morning that I held an infant in a hospital during the middle of one of the hottest summers in recent memory. But even the most thorough of descriptions won’t tell you the effect those interactions had on me and the course of my life. Description is very different from understanding function.
“The next stage in the field is to really get into function of these interactions by manipulating them,” says Gerd Blobel, professor of pediatrics at the University of Pennsylvania. “To try to alter these processes in a very specific manner will allow us to really understand what the function of these interactions are beyond the stage of description, and that’s where this field is headed”
Recently, researchers interested in uncovering the secrets of chromatin interactions have begun overlaying high-throughput data sets of epigenetics marks, revealing that the interactions of different chromatin regions are somehow linked to histone modifications and DNA methylation. This marks a major shift in the field as scientists move beyond simply describing these interactions and instead work on determining their significance in terms of cell function.
Ekaterina Khrameeva has recently returned home from a one-month stay at the lab of Leonid Mirny at the Massachusetts Institute of Technology (MIT). Mirny and colleagues use polymer physics, molecular dynamics, and high-throughput chromatin conformation capture (Hi-C) to describe chromatin folding. But they have limited experience in figuring out is the function of this folding. “So that’s why they collaborate with us,” says Khrameeva. “They want to find something functional from their Hi-C data.”
At the Institute for Information Transmission Problems (IITP) at the Russian Academy of Sciences in Moscow, Khrameeva is a junior researcher working under former Howard Hughes Medical Institute fellow Mikhail S. Gelfand. Previously, she had studied how RNA secondary structure is involved in alternative splicing, another majorstructure-function relationship in molecular biology. But now her work is focused on characterizing long-range chromatin interactions by overlaying a multitude of data sets.
For example, in a paper published in the Public Library of Science ONE last year (1), Khrameeva, Gelfand, and colleagues combined Hi-C data on the 3-D structure of the genome with data from the Encyclopedia of DNA Elements (ENCODE) project on histone modifications, DNA methylation states, and other epigenetic features. As a result, they found that chromatin regions that shared close physical proximity also shared similar epigenetic marks.
While that relationship seems to make sense, what was actually very surprising was the evidence of chimeric RNA containing exons from two or more different genes that are somehow spliced together. “People don’t know where they come from. Some people think that they are artifacts of some experimental procedure, but some groups have shown that they can be functional,” says Khrameeva.
Khrameeva and colleagues found that these chimeric RNAs are more likely to contain sequences of two genes that are physically close together. Although chimeric RNAs had been identified previously , the IITP group found them in much higher numbers than ever before and believe that they are the result of post-transcriptional trans-splicing.
Since the publication of their paper, the team has chosen a new direction for their research: topological domains. These tightly folded chromatin domains were identified in papers published last year by two different research teams (2,3).
“So what we are trying to do now, and some other groups are also trying to do, is to find the functional role of these domains,” says Khrameeva. “What I would really like to see is some systems biology approach to find the regulatory networks or some specific factors that are allocated at the borders. We see the structure, but why does it look like this? I think there should be something functional involved.”
Act Like An Enhancer
Melissa Fullwood is curious about the same thing. She was one of the researchers that helped identify those chromatin topological domains. To do so, Fullwood and her colleagues used a method that she developed during her graduate training under Ruan Yijun at the Genome Institute of Singapore called chromatin interaction analysis with paired-end tag sequencing, ChIA-PET for short. By combining the proximity ligation step of the 3C technique with chromatin immunoprecipitation (ChIP) and paired-end tag (PET) sequencing, the method can identify transcription factor binding sites as well as chromatin interactions between those sites.
“We were coming from a different idea than Job Dekker, who developed 3C and Hi-C,” says Fullwood. “Their methods grew out from trying to look at chromatin interactions and understand chromatin structure but for us, our methods came out of our interest in understanding transcription factor binding sites, particularly distal sites which might be involved in chromatin interactions to their target promoters.”
With data from the ENCODE project and their ChIA-PET experiments, Fullwood and colleagues identified what they called multiple gene clusters that were associated with RNA polymerase II (RNAPII). In a paper published in Cell, the group described how promoters at these clusters not only interacted with enhancers but also with other promoters (2). The promoter-promoter interactions were unexpected and suggested that these promoters were actually acting like enhancers.
To confirm that, Fullwood and colleagues compared their RNA polymerase II data set with a previous ChIA-PET data set for the transcription factor estrogen receptor alpha. They found that the RNAPII-bound multiple gene cluster at GREB1 partially overlapped with estrogen receptor-bound chromatin loops, suggesting that the interaction complex was partly associated with the estrogen receptor. So the researchers followed up that analysis by knocking down estrogen receptor alpha with a short interfering RNA to determine the effect on other chromatin interactions at that gene locus. Sure enough, they did indeed identify a promoter that was acting as an enhancer for other genes.
Now Fullwood is setting up shop in her new home at the Cancer Science Institute of Singapore, where she plans to investigate chromatin interactions in gastric cancer cells. Gastric cancer affects populations in Asian countries significantly more than it does Western countries. As a result, the disease is understudied. For example, more than 300 cell lines were analyzed in the ENCODE project, but not a single gastric cancer cell line was included.
So far, however, Fullwood has yet to begin performing ChIA-PET experiments on these cell lines. “As a new investigator, you don’t have the same level of resources as established investigators, and ChIA-PET is not a cheap technique.” But a new $3.6-million grant from the National Research Foundation should help to get things started, and then it’s just a matter of optimizing the method for her lab. “We need to optimize and adapt the ChIA-PET protocol to suit resources available in our lab, which may have different equipment and materials. I do not know how much optimization we will need to make the ChIA-PET protocol an everyday procedure in the lab."
Of course, she has other ideas that she would like to explore as well, such as manipulating chromatin interactions and finding the transcription factors that control these interactions. That, she says, would be the next step towards a therapeutic treatment for diseases such as cancer. “A lot of people are looking at transcription factors that are known to be important in cancer and assessing their chromatin interaction properties and studying drugs that work against these proteins to see what happens to these chromatin interactions.”
Whether It Matters
Back at the University of Pennsylvania, Blobel couldn’t agree more with Fullwood on the future direction of the field. His lab explores the genetic and epigenetic factors involved in the development of blood cells. This work includes analyzing the GATA transcription factors, which play a role in regulating the expression of adult hemoglobins. With the discovery of chromatin looping at the beta-globin locus and the development of 3C technology, Blobel ventured down the rabbit hole of chromatin structure analysis eight years ago.
When a normal blood cell develops into a adult cell, GATA1 displaces GATA2 at a gene called Kit, which is then downregulated in the mature cell. In a paper published in Molecular Cell in 2008, Blobel and his team reported that this displacement changes the chromatin interactions at this gene as well, suggesting that the chromatin structure could be involved in repressing gene expression (4).
At that time, there was a theory that chromatin looping was just a side effect of the genes and promoters binding to polymerases. Blobel and others believed that it actually might be much more involved with gene regulation than previously thought. “But we realized that when we were doing all of these 3C experiments, we just kept describing these chromatin interactions. However, this did not address whether these are cause or consequence of gene regulation,” says Blobel.
Over the next few years, Blobel and his team had a new goal: the generation of chromatin loops at an endogenous gene locus. First, the group attempted to use drug-induced dimers of transcription factors to control chromatin looping, but it did not work efficiently. “It was perhaps too simplistic to think that a single transcription factor pair can generate a specific long range chromatin loop,” says Blobel. After about 160 experiments to screen for the right conditions and dimers, Blobel was ready to give up.
And that’s when his student Wulan Deng suggested trying one more factor, namely LDB1. This factor is dependent on the presence of GATA1 at the beta-globin promoter site and is required for the proper development of red blood cells. So Blobel and Deng used zinc fingers to tether LDB1 to the beta-globin promoter, without GATA1, to see if LDB1 alone could restore the loops. In a paper published in Cell last year (5), they reported that LDB1 did indeed restore looping at the promoter site by 80% and transcription of the gene by 20%.
Having the right looping factor was everything, and the clue to use LDB1 came from the ENCODE project. Blobel and several other labs contributed profiles of nuclear factors during various stages of differentiation to the project. It was these ChIP profiles of transcription factors and histone marks that led them to the idea that LDB1 was dependent upon GATA1. “I think some people have been very critical of the ENCODE work,” says Blobel. “But it helped us here. Yes, it was costly, but people would have done this anyway little by little. I think it was nice to have done it in one nice swoop.”
Now Blobel and his group are exploring the therapeutic potential of loop formation. Instead of simply inserting genes into embryonic stem cells or progenitor cells, they will attempt to insert looping factors to generate specific enhancer-promoter interactions to increase or possibly silence gene expression. And as they do so, they’ll also explore other genome editing tools such as transcription activator-like effector nucleases and, perhaps, the CRISPER-Cas9 system, which is looking promising in its early days, according to Blobel.
“The state of the field right has been mostly at the descriptive level,” says Blobel. “People say this promoter and that enhancer form looped interactions in the genome without knowing to what extent these interactions matter.” And to do that, researchers need to start manipulating these chromatin interactions.
And for the record, that woman that I met in the bar eventually became my wife, and that infant I held in my hands was my newborn son. And the effect that both have had on my life, well, is indescribable.
1. Khrameeva, E. E., A. A. Mironov, G. G. Fedonin, P. Khaitovich, and M. S. Gelfand. 2012. Spatial proximity and similarity of the epigenetic state of genome domains. PLoS ONE 7(4):e33947+.
2. Li, G., X. Ruan, R. K. Auerbach, K. S. S. Sandhu, M. Zheng, P. Wang, H. M. M. Poh, Y. Goh, J. Lim, J. Zhang, H. S. S. Sim, S. Q. Q. Peh, F. H. H. Mulawadi, C. T. T. Ong, Y. L. Orlov, S. Hong, Z. Zhang, S. Landt, D. Raha, G. Euskirchen, C.-L. L. Wei, W. Ge, H. Wang, C. Davis, K. I. Fisher-Aylor, A. Mortazavi, M. Gerstein, T. Gingeras, B. Wold, Y. Sun, M. J. Fullwood, E. Cheung, E. Liu, W.-K. K. Sung, M. Snyder, and Y. Ruan. 2012. Extensive promoter-centered chromatin interactions provide a topological basis for transcription regulation. Cell 148(1-2):84-98.
3. Dixon, J. R., S. Selvaraj, F. Yue, A. Kim, Y. Li, Y. Shen, M. Hu, J. S. Liu, and B. Ren. 2012. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature 485(7398):376-380.
4. Jing, H., C. R. Vakoc, L. Ying, S. Mandat, H. Wang, X. Zheng, and G. A. Blobel. 2008. Exchange of GATA factors mediates transitions in looped chromatin organization at a developmentally regulated gene locus. Molecular cell 29(2):232-242.
5. Deng, W., J. Lee, H. Wang, J. Miller, A. Reik, P. D. Gregory, A. Dean, and G. A. Blobel. 2012. Controlling Long-Range genomic interactions at a native locus by targeted tethering of a looping factor. Cell 149(6):1233-1244.