Send to printer »

Feature Articles : May 15, 2009 (Vol. 29, No. 10)

Decoding Cell Communication

Understanding the Functions and Mechanisms of Every Aspect of the Cell Nucleus Now Viewed as Critical
  • Kathy Liszewski

The intricate interplay of cellular signaling pathways provides a beehive of biologic activity in the nucleus. Understanding how these pathways function and dysfunction not only helps trace the roots of cancers and genetic disorders but also suggests new therapeutic targets. At the recently held Gordon Conference on “Signal Transduction within the Nucleus,” GEN spoke with several presenters who shared their insights into this cutting-edge field.

“Behold the awesome power of yeast genetics and biochemistry,” proclaimed Ali Shilatifard, Ph.D., investigator with the Stowers Institute, as he began his keynote address. Dr. Shilatifard, who also was a co-organizer of the conference, described lessons learned from yeast about human leukemia. “My research focuses on work related to lymphocytic leukemia and mixed-lineage leukemia (MLL).

“This childhood cancer results from translocations in the MLL gene on chromosome 11. For unknown reasons, the gene appears to break in infancy. Its translocation causes leukemia with a very poor prognosis. This gene was discovered over 25 years ago by clinicians, however for a long time we did not understand its function. Therefore, we decided to take the road less traveled and use yeast as a model system to investigate the molecular functions of the MLL homologue.”

According to Dr. Shilatifard, Saccharomyces cerevisiae expresses an MLL homologue called Set1. “Yeast has been a great model system to begin dissecting our way through this pathway. We found that Set1 is a component of a larger complex we named COMPASS (complex proteins associated with Set1). Human MLL is also found in a COMPASS-like complex. Further, we determined that COMPASS can modify histone methylation, and based on these findings it was demonstrated that the MLL complex is also a histone methylase. This gives us clues as to how to find possible therapeutic targets. For example, inhibiting or modulating methylation/demethylation could be the basis for targeted treatments.”

Lately, model systems have been ignored and sidelined by funding agencies in favor of translational studies, noted Dr. Shilatifard. “I think some people do not understand the absolute power of a model system. Sarah Palin, for example, asked why should we study fruit flies. However, Drosophila, yeast, zebrafish, and other simple model systems all have the capacity to provide us with much-needed information.”

Chromatin Remodeling Complexes

Within a cell nucleus, chromosomal DNA is tightly wound and intertwined with proteins known as histones and assembled in histone/DNA units called nucleosomes, similar to beads on a string. These strands are collectively known as chromatin. “It is the chromatin that needs to be modified and dealt with, not naked DNA, to allow access by transcription factors and DNA repair machinery,” said Xuetong Shen, Ph.D., associate professor, department of carcinogenesis, M.D. Anderson Cancer Center.

Dr. Shen’s laboratory focuses on understanding the functions and mechanisms of a family of novel chromatin-remodeling complexes called INO80. “Basically, INO80 moves the beads around to expose DNA. Traditionally, people thought of chromatin remodeling primarily as a means for transcription, but we now see that chromatin remodeling is involved in DNA repair as well. When not repaired properly, cancer may result. This provides us with a potential new target for the detection, prevention, and treatment of cancer.”

The next goal of Dr. Shen’s group is to further dissect and understand mechanistically how chromatin remodeling functions in maintaining genome integrity. “We expect to use whole-genome analysis to determine where the INO80 complex binds and how it functions to prevent DNA damage,” reported Karina Falbo, a Ph.D. candidate in the Shen laboratory. “In the long term, we will also be pursuing the link to cancer by developing a mouse model to assess its possible role in causing cancer.”

According to Dr. Shen, one of the take-home messages of this work is “that we are just beginning to realize that chromatin carries a lot of information. Chromatin remodeling is likely to be involved in epigenetic changes. Understanding the function and mechanism of how these players exert their influence in the nucleus has important implications for our understanding not only about chromatin but also about the basic functions within the nucleus itself.”

Chromosome Crosstalk

The architectural organization of the nucleus is poorly understood. However, some active genes are repartitioned from the interior to the periphery of nuclear territories. “A number of studies have shown that there are interchromosomal interactions that provide a novel control mechanism to regulate gene expression,” noted Xiang-Dong Fu, professor, medicine/cellular and molecular medicine, University of California at San Diego (UCSD).

“New and intriguing questions have emerged from the recent genome-wide analyses of DNA binding sites for transcription factors,” explained Dr. Fu. “We see that there can be numerous remote chromosomal binding sites that communicate with their putative target genes by long-distance intrachromosomal and likely interchromosomal interactions.”

Dr. Fu and his collaborator, Michael G. Rosenfeld, M.D., an investigator for the Howard Hughes Medical Institute at UCSD, are working to define how such interactions take place. Their studies target genes regulated by nuclear receptors such as the estrogen receptor. “Our groups found 17β-estradiol(E2)-induced interactions between gene loci located in different chromosomes, in particular the TFF1 gene on chromosome 21 and the GREB1 gene on chromosome 2,” Dr. Fu reports.

Using fluorescence in situ hybridization, Drs. Fu and Rosenfeld found that these interchromosomal interactions appear to be mediated by a nuclear motor system. Dr. Fu added, “this is still poorly understood at this point, but it’s important to note that these gene-gene interactions are closely correlated with enhanced gene expression in response to hormones.”

Drs. Fu and Rosenfeld also determined that the interchromosomal interactions occurred in hubs called interchromatin granules. “These granules, also called nuclear speckles, are enriched with several key transcriptional elongation factors, chromatin remodeling complexes, and essentially all factors needed for pre-mRNA splicing.”

These findings underscore the ability of chromosomes to move to and interact at these common hubs, Dr. Fu noted. “For hormone-induced genes, such interchromosomal interactions in the interchromatin granules may play an important role to coordinate and enhance regulated gene expression by allowing efficient coupling of transcriptional initiation, elongation, and RNA-processing events.”

Location, Location, Location

Where a gene is positioned (even in nuclear real estate) can have major ramifications on its activity. The role of the spatial organization of chromatin within the nucleus has been debated for years. But new research has begun to shed light on the subject, according to Jason Brickner, Ph.D., assistant professor, department of biochemistry, molecular biology and cell biology, Northwestern University.

“When many genes are activated, they relocalize from the nuclear periphery, where they probably associate with the nuclear lamina, to the nuclear interior,” he said. “We are also seeing that a number of genes are specifically targeted to the periphery when they are activated. Surprisingly, if these genes are then turned off, they can remain at the periphery through several cell divisions and possess a memory of the previous activation.”

Dr. Brickner, who uses yeast as a model system, said such recruited genes can subsequently become more rapidly reactivated when the need arises. “We found that a number of inducible yeast genes are stably targeted to and remain at the nuclear periphery after they have been repressed again. For example, retention of the GAL1 gene (that encodes the galactokinase gene) lasts more than seven generations. While it is at the periphery, it can be reactivated more rapidly than the naïve state of the gene. This type of transcriptional memory requires both chromatin-remodeling factors and the histone variant H2A.Z. So, peripheral localization might represent a novel epigenetic mechanism for transcriptional control.”

The mechanism for this peripheral association relies on interactions between the nuclear pore complexes and either mRNA or DNA sequences, Dr. Brickner noted. “It’s possible that the genome may encode for its own spatial organization.”

Some intriguing and important questions remain. For example, how are cytoplasmic factors, chromatin remodeling, and gene localization coordinated to mark a promoter for rapid reactivation? How many genes use this type of transcriptional memory? Finally, what other organisms utilize this system? Dr. Brickner said “one of the most exciting possibilities is that metazoan cells might use this type of regulation. If so, this could provide a means by which environmental factors or physiological signals affect gene expression long after a stimulus is encountered. Although it’s a fantasy now, in the future we may be able to use such knowledge to therapeutically manipulate transcription.”

Hope for Hopeless Disease

Acute myeloid leukemia (AML) is another disease with a poor prognosis—only 30% of patients survive five years. “This disease can be so bad that treatment guidelines recommend clinical trials as the first option for nearly all patients with AML over the age of 60,” reported Katherine L.B. Borden, Ph.D., professor, department of pathology and cell biology, University of Montreal, and an investigator at the Institute for Research in Immunology and Cancer.

Dr. Borden and her team focus on understanding and treating the molecular basis of cancer, in particular AML. Their work has led to the development of a therapy based upon the inhibitory factor and a potent oncogene called eukaryotic translation initiation factor or eIF4E, also called 4E for short.  

“Our goal is to understand the nuts and bolts of how 4E works. Its traditional role is to initiate translation by binding the 5´ m7G cap found on mRNAs. However, elevated levels of 4E are associated with increased mRNA export. Elevated 4E levels lead to oncogenic transformation in cell culture, tumorigenesis in mouse models, and a poor prognosis in AML and other cancers.”

To study 4E, Dr. Borden isolates leukemia cells from the blood of patients. “Our studies found that 4E also governs cell-cycle progression and cellular proliferation by orchestrating the expression of several key genes at the post-transcriptional level. We found that a commonly used drug called ribavirin impedes 4E’s ability to make cells cancerous without significantly affecting normal cells. We are targeting this in the clinic. We found that ribavirin was a natural mimic of 4E’s ligand and shuts down its transforming and apoptosis rescue functions. We had found this in patient specimens previously (for those with highly elevated 4E levels)  and have concluded a Phase II trial with ribavirin.”

“We had previously seen that ribavirin inhibits growth in cell culture of patient specimens. The Phase II trials determined that ribavirin inhibits 4E function in vivo and that also correlates with response. There are still some challenges, however. Patients can develop resistance. So we are looking now at more targeted therapy such as combining ribavirin with chemotherapy. We’ll also perform high-throughput screening to find other chemicals that could have synergy with ribavirin.”

Genome Organizer

Metastasis of cancer cells is the final step in solid-tumor progression, and the most common cause of death in cancer patients. One culprit implicated in an aggressive form of breast cancer is special AT-rich sequence binding protein 1, or SATB1.

“SATB1 is a nuclear protein that normally plays a critical role in regulating gene expression during thymocyte differentiation and activation of T cells,” said  Terumi Kohwi-Shigematsu, Ph.D., scientist in the life sciences division of the DOE’s Lawrence Berkeley National Laboratory. “However, in breast cancer cells, once SATB1 becomes expressed, it reprograms expression of a multitude of genes to promote tumor growth and metastasis.”

Dr. Kohwi-Shigematsu said that SATB1’s role in breast cancer represents a new paradigm. “We believe this is a new model of gene regulation leading to tumor progression. A key question we are investigating is how SATB1 alters expression of so many genes. In our studies, the expression of more than 1,000 genes is altered by SATB1 expression in breast cancer cells. SATB1 in the nucleus has a unique architectural distribution, onto which its target genes are anchored and assembled with chromatin-modifying enzymes and transcription factors, thus providing a regulatory network.

“Therefore, such a regulatory network must play a critical role in regulating the epigenetic status of chromatin and gene expression. Similar to the case found in activated T cells, it is also likely that in breast cancers, SATB1-targeted genes collect into intra- or even interchromosomal loci to form chromatin loops so that they can be coregulated. What we are seeing is an emerging link between chromatin remodeling enzymes, epigenetics, and cancer.”

Understanding the mechanisms used by SATB1 could provide a new diagnostic and prognostic marker as well as a therapeutic target for breast cancer. “Most companies are interested in targeting proteins on the cell surface. In the future, however, an effective strategy might be developed to target proteins in the nucleus such as SATB1. By discovering which signaling pathways are important for SATB1 activation we may be able to target them for therapeutics as well,” Dr. Kohwi-Shigematsu concluded.