Chemical genetics enables an approach to life phenomena and developments in drug discovery
10 October 2008 (Volume 3 Issue 10)
Chemical biology is a study that uses chemistry to explore life phenomena. Chemical genetics is a part of chemical biology in which scientists arrange for compounds such as microbial metabolites to act on cells to investigate the proteins that they act on and the changes that occur, thereby establishing the relationship between genes and their functions. There is a good possibility that chemical genetics will be able to explain complex life phenomena, and also that compounds produced by microorganisms will lead to the discovery of new therapeutic agents. At RIKEN, scientists have started a full study of chemical genomics, large-scale chemical genetics that targets the whole genome. This article reports on what is happening at the forefront of chemical genetics and chemical genomics, which are currently hot topics for both basic and applied research.
I was a dropout as a natural products chemist!
“What did you have for breakfast this morning?” asks Minoru Yoshida, Group Director. You may have had Japanese food such as natto, a traditional fermented soybean food, and miso soup; or western food such as bread and yogurt. “Natto, miso soup, bread and yogurt are all produced by microbial fermentation, and so are soy sauce, cheese, pickles, alcoholic beverages, penicillin (an antibiotic) and mitomycin (an anticancer drug). Microorganisms can produce various substances, which are all useful. Thus, a stone monument named ‘Microbe Mound’ has been built to express thanks to microorganisms (Fig. 1).”
Figure 1: Microbe Mound.
The Microbe Mound is in the holy precinct of the Manshuin Temple in Kyoto. The title engraved calligraphically on the stone monument was written by Kin-ichiro Sakaguchi, Professor Emeritus at the University of Tokyo, a world authority on applied microbiology and the first Vice-President of RIKEN when it was a government-affiliated company. He was also known as ‘Doctor Sake’. Yoshida studied at the Laboratory of Fermentation and Microbiology in a direct line to Sakaguchi at the Department of Agricultural Chemistry in the Faculty of Agriculture at the University of Tokyo. “I started my studies as a natural products chemist trying to find new compounds produced by microorganisms.”
While studying for his doctorate, Yoshida discovered trichostatin A in an actinomycete culture solution. “Investigation showed that it was a known substance. In natural products chemistry, much importance was placed on finding new products. Therefore most researchers do not turn to known substances. However, I was so interested in trichostatin A that I could not bring myself to look for other new compounds. I was a dropout as a natural products chemist,” says Yoshida with a smile, looking back on those days.
First of all, natural products chemists investigate the structure of compounds, or their ‘faces’. “In most cases, we can find out whether or not a newly discovered compound has useful functions by comparing its structure with that of known compounds.” While Yoshida was investigating trichostatin A, he realized that it has a very unusual structure. Trichostatin A was administered to mouse leukemia cells, and surprisingly the cells changed to normal erythrocytic cells. It was also proved that trichostatin A is effective at a concentration of only 10 nM. “We decided to use what is now called ‘chemical genetics’ and started investigating which proteins trichostatin A acts on, and how it works on them.”
What is chemical genetics?
Genetics is a study that relates genes to phenotypes, which are eventually observed as forms or characteristics. In genetics, scientists try to find mutant strains that are variations of phenotypes, and identify the genes responsible. Recently 'reverse genetics' has become very popular, in which researchers select interesting genes from among those whose nucleotide sequence, encoding the genetic information, has been determined, and cause the selected genes to mutate so that they can investigate how the phenotypes change. Phenotypes of higher organisms such as human beings, however, may not change because they have multiple genes with the same functions; one gene can therefore compensate for another if one of the genes mutates.
Thus, chemical genetics has offers a way of overcoming the problem. “When compounds are administered, they bind to some special proteins and inhibit their functions, even when they are produced from different genes with the same function, inducing similar phenotypic changes to those caused when genes are mutated. Thus, chemical genetics is a study that takes advantage of compounds to clarify the relationship between genes and phenotypes,” explains Yoshida.
Genes with sequences of the same kind can produce proteins of the same kind. Because compounds can act on all proteins of the same kind, no functions are complemented by different gene products with the same function. In this way, phenotypic changes can be investigated. In addition, the changes occur only when compounds are being administered, and they are easily observed because phenotypes return to their original state when the compounds are removed. Unfortunately it was difficult for conventional methods based on genetics to investigate the functions of essential genes because they destroy genes essential for survival and stop embryonic development. In contrast, the use of compounds can provide a method for investigating essential genes.
Chemical genetics is attracting greater attention as a new area of genetics. Yoshida says, “This is because the approach based on compounds could lead to the discovery of many interesting and unpredictable phenomena, and it is hoped that the compounds will lead directly to therapeutic agents.”
Toward therapeutic drugs to treat diseases such as cancer and neurodegenerative diseases
The application of chemical genetics to trichostatin A, which was what Yoshida was really interested in, revealed that it binds to a protein known as histone deacetylase (HDAC) (Fig. 2).
Figure 2: Inhibition of histone deacetylation by trichostatin A.
A strand of DNA winds around histone proteins. When histones are acetylated by a protein called histone acetyltransferase (HAT), the strand begins to loosen, which allows gene expression. After a short time, the histones are commonly deacetylated by HDAC, which causes the gene expression to stop. Trichostatin A binds to HDAC and inhibits its function. As a result, the histones remain acetylated and gene expression is activated. Trapoxin works in the same way as trichostatin A.
Histones are proteins around which a strand of DNA is wound, and HDAC can remove acetyl groups from the histones. Trichostatin A binds to HDAC proteins and thereby inhibits histone deacetylation, producing a stable condition in which genes are easily expressed.
The discovery of a specific histone deacetylation inhibitor itself is an amazing event because no such inhibitors had previously been discovered. A further surprise comes as Yoshida says, “I do not know why, but when trichostatin A inhibits the function of HDAC, the expression of genes that serve to cure diseases is activated, not the expression of genes that cause diseases.” At present, besides trichostatin A and trapoxin, many compounds that inhibit the function of HDAC have been discovered or synthesized, of which about ten have been clinically tested as anticancer drugs. Yoshida continues, “In addition to anticancer drugs, HDAC inhibitors are attracting increasing attention because they are reported to be effective in treating nerve-cell diseases such as Huntington’s disease and Alzheimer’s disease.”
Splicing inhibitors aimed at anticancer drugs
The following describes Yoshida’s recent findings. He focused on FR901464, a compound known to have anticancer activity against colon cancer cells and lung cancer cells, and conducted an experiment using spliceostatin A, which is a modified compound of FR901464 that has been slightly changed in structure for stabilization. “I thought that spliceostatin A would bind to HDAC to inhibit histone deacetylation in the same way as trichostatin A. To our surprise, however, we found that spliceostatin A binds to SF3b, a protein complex necessary for splicing, not to HDAC (Fig. 3). In other words, spliceostatin A is a splicing inhibitor.”
Figure 3: Spliceostatin A and the function of its target SF3b.
Spliceostatin A inhibits splicing by binding to protein complexes called SF3b. As a result, intron-containing immature mRNAs accumulate in the nucleus. Some immature mRNAs also move out of the nucleus and are then translated into abnormal proteins. It has been shown that SF3b is not only essential for splicing but also has the function of keeping immature mRNAs within the nucleus by binding directly to the introns.
Splicing is an essential process in the production of normal proteins: it removes the non-coding ‘introns’ from the RNA that is transcribed from a DNA sequence, and correctly joins up the remaining coding ‘exons’.
At the same time, Eisai Co. Ltd., a pharmaceutical company, successfully found another compound that binds to SF3b and inhibits splicing, demonstrating that it served as an anticancer drug. “It is natural to think that if splicing is inhibited, no normal proteins will be produced, and therefore normal cells will die. Why, then, do cancer cells die selectively? It is a mystery. In the future, we will collaborate with Eisai to figure out what is behind this unexplained phenomenon. Therapeutic agents that can inhibit DNA replication, transcription, and translation into proteins have been developed, but no splicing inhibitors. We hope that our study will lead to the development of anticancer drugs that work on a new mechanism.”
Additionally, it was found that when splicing is inhibited, some immature RNAs accumulate in the nucleus, and others move out of the nucleus. The immature mRNAs with introns are then translated into abnormal proteins, although under normal circumstances the introns would be removed. “SF3b is not only essential for splicing, but it also has the function of keeping immature mRNAs within the nucleus by directly binding to the introns of mRNAs.” The intron has recently become one of the important subjects of research in life science.
“Researchers are now starting to say that introns have unique functions although previously they were considered redundant areas. Normally, introns are not stable in a nucleus. However, their functions can be investigated by using SF3b inhibitors, which make introns accumulate within the nucleus in large quantities. These findings may be the beginning of ‘intron biology’, in which intron functions are explored.”
In addition to trichostatin A and trapoxin, which inhibit histone deacetylation, and spliceostatin A, which inhibits splicing, Yoshida discovered leptomycin B, which inhibits the nuclear export process in which proteins are transported from a nucleus to the cell’s cytoplasm. He is advancing his own research using these newly discovered compounds (Fig. 4). All of them are expected to lead to the development of therapeutic agents such as anticancer drugs.
Figure 4: Control mechanism of a cell, which has been clarified by chemical genomics.
Yoshida has used compounds produced by microorganisms and modified versions of the compounds to identify target proteins. As a result, important cellular mechanisms such as gene expression and transport of proteins have been clarified one by one. The figure illustrates the normal functions of a cell. Each compound binds to a particular protein, inhibiting normal cell functions.
Infrastructure development for chemical genomics
“Conventional chemical genetics has been studied with unexpectedly discovered compounds that cause interesting phenotypic changes. The future is the time of ‘chemical genomics’, a large-scale version of chemical genetics targeting the whole genome in an effort to find both useful compounds and their targets by applying them to a variety of genes,” says Yoshida. Chemical genomics, however, requires sophistication as regards both chemistry and biology. The current situation is that chemical genomics is not expanding as expected because of a shortage of researchers who have a good knowledge of both chemistry and biology. Yoshida says, “We are establishing a special system such as the ‘Localisome’ database, using fission yeasts so that every researcher can study chemical genomics” (Fig. 5).
Fission yeasts are model eukaryotic organisms that have many genes in common with humans. The complete genome of fission yeast was decoded in 2002, in which 4,948 genes were found that can produce proteins. Yoshida extracted 4,910 genes from the yeast (about 99%), and successfully produced proteins. He also tried to attach fluorescent labels to proteins so that he could establish ‘Localisome’, a database showing in which part of the fission yeast the protein is located. When a compound is administered, it changes the localization of proteins. The protein related to the compound can be identified without special knowledge by comparing it with the basic data in Localisome. The establishment of Localisome was a six-year project. “If our purpose had been limited to the establishment of the database, we would never have engaged in such laborious work. We pushed hard because we were seeing the future direction of the contribution to drug discovery.”
Figure 5: Comprehensive analysis of fission yeasts that serves as a basis of chemical genomics.
The Chemical Genomics Research Group successfully cloned about 99% of all the genes that produce proteins of fission yeasts, and is now establishing a system that can exhaustively analyze the cloned genes. It is easy to tell which proteins change their gene expression levels, mobilities in gel electrophoresis, or intracellular localization when compounds are administered, by referring to databases such as ‘Localisome’, a database related to the localization of proteins in a cell, and ‘Mobilitome’, another database related to the electrophoretic migration positions of proteins.
A Chemical Genomics project was started in the US in 2004, mainly at the National Institutes of Health (NIH) Chemical Genomics Center. In Europe and China, they are advancing the infrastructure development for chemical genomics. In the past, natural products chemistry was known as ‘Japan’s specialty’. However, it has to be admitted that Japan lags behind in the field of chemical genomics. Therefore, in April 2008, a new Chemical Biology Department was established in the Advanced Science Institute at RIKEN. “Unique compounds are a starting point for both chemical biology and chemical genomics. No further progress can be made without them. We are trying to enhance the research infrastructure by focusing on the chemical bank that Hiroyuki Osada, Group Director, is promoting.” Thus, they are advancing research with a view to contributing to drug discovery.
Yoshida asserts that he is most interested in ‘epigenetics’. Living organisms are not always governed by the genetic information encoded in DNA. It is known that the function of newly translated proteins is changed when they are acetylated, methylated, or ubiquitinated. Epigenetics is a process in which the changes in gene functions caused by the acquired modification of proteins and DNAs are passed on to descendant cells for generations. “Regulation of gene function by histone deacetylation is a perfect example of epigenetics. Thus I think I will continue to be involved in the study of trichostatin A in the future.” The study is likely to lead to the development of therapeutic agents such as drugs for the treatment of cancer and neurogenerative disorders such as Alzheimer’s disease.