|
||||||||
|
TOXICOLOGY Chip Data Show Genetic Ups and DownsMany more reviews and commentaries have been written on how DNA chips will irrevocably change biology, medicine, and our daily lives than papers have been published with any data to prove it. It is the talk of the biotech industry and talks on it jam conference rooms. There is not much to these chips, just small pieces of glass with an imperceptible grid. What that grid can hold, though, is the entire genome of an organism, the full complement of genes that form the how-to manual for building the worm C. elegans--or a human being. And what these DNA microarrays, or DNA chips, can measure is the varied gene expression of cell, organ, or organism when developing, growing, dying, dividing, being irradiated or infused with chemotherapy, warmed or cooled, poked or prodded. Getting a readout of gene expression with any manipulation is possible--but what does any of it mean? Leona Samson and Scott Jelinsky, both of HSPH, are beginning to learn. As authors of one of the few published papers on the use of microarrays, they have seen the road ahead and discovered some surprises along the way. Samson's lab is interested in how cells respond to alkylating agents, chemicals that can cause both specific and general lesions in DNA. "Alkylating agents are found in combustion products, cigarette smoke, in food substituents, endogenously inside cells as normal metabolites--and a lot of chemotherapeutic agents are alkylating agents," says Samson, professor of toxicology. "What are our natural defense mechanisms against these agents and how do we prevent killing mutations? There are many mechanisms that help cells prevent, repair, or recover from damage of DNA," she says. One cellular response to a toxic agent is to bump up transcription of genes that will protect the cell. A yeast gene studied by Samson that is known to be upregulated in response to DNA damage is the glycosolase Mag1, whose protein is directly involved in excising a damaged DNA base. Traditionally, a single gene such as Mag1 or a small handful can be examined by methods such as Northern blots, and it is this level of transcriptional regulation that Jelinsky, a postdoc, came to the lab to study. Using transcript arrays was "something that fell into my hands when I got here," he says. "Our initial idea was to treat cells with chemicals that damaged DNA and look for genes that are inducible, with the idea that a majority of them would be directly involved in DNA repair." They expected a small number of genes, 80 or so, but using only one alkylating agent with one time point they found more than 400 genes of the 6,200 total yeast genes whose transcripts were induced or repressed. "At first I was very happy that we found a lot of genes, but after I started going through it, I was a little scared," Jelinsky says. Besides finding known DNA repair genes, Jelinsky found genes involved in signal transduction, cell wall synthesis, membrane transport, protein degradation, amino acid metabolism, ribosomal RNA synthesis, and others. "It could be that many of these genes have nothing to do with how cells recover from alkylation damage, or we may have discovered a whole host of new genes that help cells recover from alkylation damage. It's our job to figure out which is true," says Samson. One transcript array experiment is only the beginning, and the directions researchers can take with such a mass of data are incredibly varied. One way to figure out which genes are truly involved is to use the fully sequenced genome of yeast. In collaboration with the laboratory of HMS professor of genetics George Church at the Lipper Center for Computational Genetics, Jelinsky is using the yeast genome database to find common regulatory regions of the genes that are coordinately up- or down-regulated. Then, using literature searches or techniques such as a one-hybrid assay, they hope to find the transcription factors that regulate these groups of genes, or regulons. With this data they can begin to knock out regulation of small groups of genes involved in the alkylation response. Similarly, genes whose involvement is unclear can be knocked out and the strains tested either for increased sensitivity to DNA damage or for an altered transcript profile. In this way the biology of DNA damage and repair can be teased apart. Another direction these types of experiment could lead is into the interdisciplinary field of toxicogenomics, which uses transcription profiles of an organism or cell treated with a toxic chemical as a biological fingerprint for that chemical. If an unknown compound has a similar transcription profile to known compounds, it may have similar biological effects. Whether these first experiments lead to a better understanding of how cells respond to different chemical assaults or are used to define how previously undescribed chemicals fit into a set of known responses, the flood of data from microarrays has only just begun. --Justin Yarrow |
