Analysis of the eye developmental pathway in Drosophila
using DNA microarrays.
Summary and Critique
Pax-6 genes encode transcription factors that are capable of activating an eye development pathway that appears to have been evolutionary conserved for both vertebrate and invertebrate species. The Drosophila Pax-6 gene analyzed in this experiment was eyeless. Eyeless was analyzed in the third larvae stage, when the cells are starting to differentiate in the photoreceptors. Two different full-genome microarrays were used to analyze eyeless. Comparing gene expression in wildtype leg discs to leg discs where eyeless was ectopically expressed made it possible to identify 371 genes. These 371 genes met two criteria - first, the genes are expressed in eye discs and second, the genes are up-regulated when eye formation is ectopically induced. Using two different full-genome microarrays decreased the number of false positives for eyeless-induced genes.
The purpose of the authors’ experiment was to better understand the Drosophila eye developmental pathway using DNA microarrays. The authors “identified a number of previously described genes that were not yet known to be expressed during eye formation and suggest a possible role in eye development for previously uncharacterized genes (Michaut 4029).” Most of the genes induced by eyeless are “transcription factors involved in photoreceptor speciation, signal transducers, actin-binding proteins, cell adhesion molecules, and proteins involved in cell division (Michaut 4029).” The authors hope that by identifying genes induced by eyeless in the Drosophila eye developmental pathway they will be able to gain knowledge of the morphogenesis of the mammal eye through comparison.
I do not disagree with any of the conclusions in this paper. It is true that a better understanding of the specific gene pathway that eyeless induces will likely lead to a better understanding of how mammalian eyes form.
Figure 1
A) Figure 1A shows eyeless-induced genes that are detected by both DNA microarrays, the roDROMEGa and the DrosGenome. RoDROMEGa has 228 eyeless-induced genes expressed ectopically and expressed in eye. DrosGenome has 198 eyeless-induced genes expressed ectopically and expressed in eye. This figure is useful because it shows that 55 eyeless-induced genes are detected by both the roDROMEGa and the DrosGenome. Since these 55 genes are detected by both full-genome DNA microarrays, there is greater certainty that these eyeless-induced genes are found in the Drosophila eye development pathway.
B) Figure 1B defines the function of 254 of the 371 eyeless-induced genes. This figure could have been more helpful if it described the specific genes that they were categorizing instead of just giving the number of genes in each classification. However, the figure was helpful because it shows the types of genes that eyeless most abundantly induces, such as transcription factors. This helps us understand what types of genes are generally induced in the Drosophila eye formation pathway. The figure also points out that many of the details are not known about the eye development pathway since 117 of the 371 eyeless-induced genes cannot be assigned a molecular function.
Table 1
Table 1 indicates 55 genes expressed in leg and eye imaginal discs and whether their transcription increased or decreased during ectopic eye formation. The 55 genes are detected by the two microarrays according to a three-selection criteria - yellow, orange, and red shading. The different colored shadings represent the fold induction value for each individual gene for both DNA microarrays. Fold induction values measure an increase in gene transcription after eyeless has been expressed. The information provided in Table 1 shows that the two full-genome DNA microarrays of Drosophila produce different fold induction results. These different fold induction results for the two DNA microarrays explain why two DNA microarrays were used instead of one - to increase validation of gene expression.
Table 2
Table 2 describes DNA microarray values for the genes discussed in the text. The information obtained from the two microarrays sometimes differed. For example, in the DrosGenome1 microarray, no increase in the transcription of cyclin E after ectopic eyeless expression in the leg discs was detected, but transcripts were detected in the eye discs of DrosGenome1. In contrast, the roDROMEGa DNA microarray found that transcription of cyclin E in the leg discs was up-regulated after ectopic eyeless expression, but cyclin E endogenous expression was not detected in the eye discs. Since it is difficult to determine which microarray is more correct, it is important to use two microarrays to increase the validation of results. Also, as in Table 1, some genes are highlighted blue because their presence in eye discs has been validated by the use of gene expression (SAGE) tags in eye disc libraries.
Figure 2
Using in-situ hybridization, Figure 2 shows where eyeless-induced genes are expressed in wild-type eye discs. The locations of the genes were described relative to the area of the morphological furrow. This figure is particularly useful because it provides information about genes in which no molecular function had been realized. The authors used this information, as well as the information obtained from Figure 1, Table 1, and Table 2, to identify a number of characterized genes that were not yet known to be expressed during eye formation. The authors and were also able to suggest possible roles for previously uncharacterized genes. I feel that the authors did a good job describing how they used DNA microarrays to identify previously characterized genes that were not yet known to be expressed in Drosophila eye formation. However, I think that the authors did not provide much information explaining how exactly they were capable of suggesting possible roles for previously uncharacterized genes. Instead of showing data to support their suggestions for gene function, the authors merely stated their suggestions. For example, the authors write, “transcription of three previously uncharacterized genes potentially encoding cell adhesion molecules” without describing any data to support this assumption (Michaut 4028).
Future Experiments
Even though I think that the authors provided sufficient data for identifying the eyeless-induced genes expressed both ectopically and in the eye disc, they could be even more thorough by using a couple more DNA microarrays to obtain even more data to further validate the genes they have already identified in eye development or to discover genes that they might have missed.
A second option for future experimentation would be to perform tests on the genes that the authors claimed to have characterized to see if their suggested functions are correct. Knockout or knockdown methods can be used to identify gene function. Since the gene sequence for the uncharacterized genes is known, you can use the knockout or knockdown method to compare the formation of eyes with and without the presence of a specified gene. One can then observe the differences in eye formation to learn more about the function of the uncharacterized gene. The knockout or knockdown method would not only help when analyzing the genes that the authors’ claimed to have found a function for, but this method can also be used to identify the functions of the 117 genes in the authors’ paper that have no assigned molecular function.
References
Campbell, Malcom. DNA Microarray Methodology - Flash Animation. April 2003
<
Michaut, Lydia et al. Analysis of the Eye Developmental Pathway in Drosophila Using DNA Microarrays. April 2003 <http://bio.davidson.edu/courses/Molbio/restricted/2003Rev/Flyeyechips.pdf>
Wagner, Thomas. Scientists Invent Faster Gene Function Identification System. April 2003 <http://www.ohiou.edu/NEWS/months/feb98/160a.html>
Davidson College Molecular Biology Homepage
Send questions or comments to brstonestreet@davidson.edu