PerspectivesAre you interested in submitting a Perspective Article? Be sure to read The Science Advisory Board's Editorial Guides for Perspective Articles. Click here. Toxicogenomics: Do Our Expectations Exceed the Technology? by James E. Trosko, Ph.D. This article is a review of the following paper: F. Boess, et al," Gene expression in two hepatic cell lines, cultured primary hepatocytes and liver slices compared to the in vivo liver gene expression in rats: Possible implications for toxicogenomics use of in vitro systems". Toxicol. Sci. 73: 386-402, 2003. The major challenge in using DNA micro-array technology on "normal tissue" and comparing the results to "diseased or treated" tissues is that all tissues contain three types of cells ( pluri-potent stem cells; progenitor cells and terminally-differentiated cells). The stem and progenitor cells can exist in the G-0, or cell cycle stages [expressing different genes], going through apoptosis; be stressed, while the terminally differentiated cells express both their differentiated genes, stress genes, adaptive response genes]. Therefore , an extract of this "normal" tissue would contain a mixture of expressed genes in proportion to the number of the three types and the different stages each cell type might be in at the time of sampling. The disease tissue is basically the same except it the treatment or disease process could alter both the proportion of cell types and the stages of the cell cycle they might be in. This paper by Boess et al demonstates clearly that in vitro cell lines express different genes than the primary hepatocytes, in turn from the genes found in the liver slices. Clearly, cell lines are usually immortalized or transformed cells ( which , in principle, are "clonal", but by the time they reach useable densities, they have mixtures of clones of altered phenotypes or mutant clones. Also, depending on whether these cells are in log phase or stationary phase, they will be expressing different genes. In addition, depending on the substrate of the culture dish, the factors in the medium (high/low calcium, etc.), the oxic state of the cells, these cells would express different genes. In primary hepatocytes, while "normal" cells, soon lose their normal gene expression, because the traditional culture medium, oxic conditions, substrates,are not identical to the in vivo conditions of the liver. Even the normal liver slice contains the three types of cells that are going to be different from the diseased or treated liver. If a toxin or toxicant differentially affects one cell type over another cell type in vivo, one would have to determine this, not from the gene profile, but from cell biology. Of course, if one observes a consistant pattern of gene expression in vivo and associates this with a pathological response, i.e., hyperplasia, inflammatory response, this could constitute some useful information. It would take more to assign 'causation" to the pathological response, since the gene pattetrn might be more due to the consequences of the original trigger leading to the pathological response. In summary, this paper by Boess will humble all of us about the uncritical use of this powerful technology. It should help us be more critical about both the design of experiments using this technology and the interpretations of any results obtained. It sure does not mean that in vitro techniques are useless. They still can be used for "mechanistic-type" studies. This paper also provides solid evidence explaining why so many in vitro assay results do nor lead to predictable in vivo results. On the other hand, using DNA micro-array technology only on in vivo tissues leads to results which, to date, is too complex for easy interpretation of potential mechanisms of action. ### Dr. Trosko has been a member of The Science Advisory Board since August 2002. ### << Previous Next >> [ View All Perspectives ] |
|
