This web page was produced as an assignment for an undergraduate course at Davidson College.


Review of a Published Paper

Paper: Analysis of the eye developmental pathway in Drosophila using DNA microarrays

Lydia Michaut, Susanne Flister, Martin Neeb, Kevin P. White, Ulrich Certa, and Walter J. Gehring

April 1, 2003 PNS Volume 100, p. 4024-4029


Michaut et al. began researching on the development of eyes in Drosophila with the knowledge of Pax-6 genes. This set of genes are known as transcription factors, which play the role of activating other genes. The Pax-6 genes of interest here turn on secondary genes that code for the development of eyes. The presence of one specific gene, known as eyeless (abbreviated ey), codes for one of two Drosophila Pax-6 genes that activate the complete set of genes for eye formation. However, when ey is found in other discs than where normal eye formation occurs (in this paper the ey gene is manipulated in the leg discs), then ectopic eyes can physically develop in various locations. In this paper, the goal was to highlight genes involved throughout the entire eye morphogenesis. Research involved the use of DNA microarrays to compare wild-type leg discs to others where ey is ectopically expressed and from there analysis of endogenous expression of other genes found in eye discs occurred to lead to the hypothesis of the cascade of events (genes activated by set signals) in the development of eyes.

Gene expression was examined by obtaining RNA from a given tissue and then hybridizing such RNA (now labeled) into the microarray. DNA microarrays are small chips consisting of many different oligonucleotide sequences that constitute the genes found throughout the organism (Campbell, 2002). Once hybridized, colored spots that may appear on the DNA microarray serve as positive signs that genes in the location of the colored spot are indeed expressed. In this paper, researchers wanted to see which genes were expressed via an induction from the ey gene, so they utilized a comparison between RNA obtained from normal leg discs, leg discs with ectopic ey, and normal eye discs. The normal eye discs and the leg discs with ectopic ey expressed genes that the normal leg discs fail to color in the DNA microarray. These genes found in the normal eye discs and the leg discs with ectopic ey were believed to be genes activated by ey in the development of eyes, but for confirmation, these genes were analyzed further.

Research continued by using a second microarray, so as to rule out any experimental errors in the genes expressed. Results were then obtained from the two microarrays and led to the identification of 371 genes activated by ey (figure 1A). Of those 371 total genes, only 55 were seen in both microarrays for normal eye discs and leg discs with ectopic expression of ey; these 55 genes were not expressed in normal leg discs (figure 1A). Of these 55 genes, most had already been researched and their function understood (figure 1B). In figure 1B it is clear that most of these activated genes served as enzymes, transcription factors, and signal transducers. From there, a look at table 1 which represents the function (where known) and statistical value of each of the 55 genes in both microarrays from the eye discs and ectopic eye formation in leg discs via ey, enabled researchers to make another conclusion.

Researchers determined that some genes which had previously been deemed necessary for eye formation did not occur within their 55 genes. The conclusion, as alluded to in the previous paragraph, was that certain genes are specialized in the development of the retina and are therein seen in that specific cascade of events, as opposed to in the overall formation of eyes as physical structures. Following this discovery, researchers noted several transcription factors and signal transducers that were not tied to eye development, but had other signal pathways affiliated. More specifically, 38 total transcription factors were seen induced by the ey gene and in normal eye discs; table 2 reflects the results from such transcription factors through the use of the microarrays. Of these 38, 18 transcription factors were already associated with eye development and even more research had been done to prove that these transcription factors are associated with photoreceptor differentiation. There was also 3 other genes that were noted to function at the beginning of the eye development cascade of events. The text goes into detail of how many of the genes in table 2 (actually seen in tables 1 and 2) were confirmed using serial analysis of gene expression (SAGE). The underlying strategy of this process is to synthesize short cDNAs, referred to as tags, from all the mRNAs in a given cell and then link the tags together in clones that can be sequenced to learn information regarding the nature of the tags, and therein the nature of the genes expressed in the cell, and finally the extent of expression of such genes (Weaver 800).

Towards the end of the paper, attention is brought to the location of the genes that have been discussed earlier (those involved in the morphogenesis of the eye in Drosophila) (more specifically the genes in table 1). The different genes are highlighted, withn their respective developmental discs (figure 2). In a sense, this is the culmination of the research because the researchers are able to physically reveal, via in situ hybridization using digoxigenin-labeled antisense RNA probes, where the different genes act in the development of the eye. Genes playing both known and unknown roles were identified in this paper and at this point those involved in eye formation are clear.

In the end, the overall conclusion drawn in this paper is that ey primarily activates genes that play a role in early retinal differentiation. The genes that are thereby induced for all of these processes to occur include: trancscription factors involved in photoreceptor specification, signal transducers, actin-binding proteins, cell adhesion molecules, and proteins affixed to cell division. The ey gene turns these genes on and is thereby established as a master control gene for the morphogenesis of eyes in Drosophila. Additionally, the role of other genes is noted along the reseach pathway for the ey gene.


Critique of this Paper

This paper does a fine job of adding validity to the overall research performed. The majority of the validity stemmed from the use of two separate microarrays. The two microarrays serve to compare data obtained under two different backgrounds and analyze the gene expression in both (backing up the findings in one by looking at the expression in the other as well). Through the textual explanations of the need for the two separate microarrays, as well as descriptions of the overall research methods utilized beyond just the microarrays themselves, the researchers relinquish doubt of findings that could be present in the mind of the reader. The only problem that arose midway through the paper, concerning these two different microarrays, was that some conclusions were made just through analysis of one microarray. For example, certain genes were expressed and then researchers were able to develop hypotheses as to how they function in the development of the eye. These new findings were exciting and noteworthy as the initial glance. However, the problem was that the paper boasted of finding 55 genes detected by both microarrays and then went on to discuss new findings on those genes that were only seen only in one of the microarrays. It is as if the researchers pride themselves on the use of two different systems, but then fall back to provide much discussion on what just one of the systems revealed. A case can be made for the fact that some of these genes have already been researched and known to be integral parts in the development of the eye and were also noted by SAGE (not just in the microarrays), but it was questionable how the discussion of findings did not solely rest on the total 55 genes that both microarrays expressed.

Another aspect of the paper which had both positive and negative aspects was the wealth of statistics. While research undeniably relies on statistical information, it was unusual to see such a large amount of information gathered using statistics in a molecular biology publication. Perhaps this is an incorrect assumption, but from the majority of other molecular biology research perviously examined, one is led to believe that statistical tests such as p- and t-tests are typically reserved for field studies. While it does not withdraw from the validity of the research, it was a different twist on data obtained, in a way. So while the types of mathematical tests utilized were clearly explained in the paper, the overall need for them was not completely justified. Of course it never hurts to include the maximum amount of data or research possible, this paper did not point out how the computer figured out the statistical numbers and the overall feeling obtained was a sense of doubt in the need or relevance that could be placed on the stats dropped into the text alongside figures and tables.

Finally, the in situ hybridization would have generally been an effective means of displaying the genes, but in this paper it failed to completely do so. The bands that were supposed to be the focus of the figure just did not boldly represent what the researchers intended to be a clear representation of the genes expressed. Granted, the overall idea of performing the in situ hybridization was effective and additionally, the idea of comparing the normal eye discs with wild leg discs and leg discs where ey was ectopically expressed, was effective as well. The argument behind this is a concrete means of demonstrating that the genes expressed in both the ectopic ey leg discs and the normal eye discs and concluding (with the two microarrays) that the genes activated in both are involved in the development of the eye. The overall use of microarrays, SAGE, and in situ hybridization, all generally added to the publication regarding how the ey master control gene functions.


Possible Future Research

The end of this paper lends itself to further research into the particular function or roles of genes that were expressed in just one of the two microarrays. There was an entire host of genes that separate experiments could delve into. More specifically, research could be performed to ectopically represent some of the genes, other than ey, in other areas to see what influence they have in comparison to the influence of the master control gene here. There was a short discussion on the genes placement along the signal cascade, so perhaps the research could begin with those genes that ey is believed to activate first in the sequence. It would be very interesting to see how far along in development the eye would form if the same experimentation occurred using ey and then perhaps terminating the expression of genes just after the next gene is expressed. This could lead to an entire set of separate experiments just on one or two other genes (or many).

Research could continue along the lines of comparing the formation of Drosophila eyes, using ey, compared to the development of mammalian eyes. If the Pax-6 genes are as conserved as previous research has led researchers to believe, then further exploration into how similar the development of eyes are in Drosophila compared to particular mammals would be very interesting. The question of differentiation, in terms of the type of eye that each animal exhibits, would be interesting to see in terms of how far along in their signal cascades the same genes are activated. There could be just a few genes that are different in human eyes versus Drosophila that establish a completely separate eye through which each organism visualizes objects in their respective environments. Within the contexts of that research to compare the signal pathway from ey onward, researchers could look for the specific binding proteins (transcription factors). This would set up another set of experiments wherein transcription factors are the focus in the devolopment of eyes. The same research techniques could be used up to this point, as well as footprinting for the specific DNA at this point. This footprinting method could see if the DNA sequence is conserved and then lead to research into the direction of the genes involved beyond the master control gene.

Generally, the research obtained up to this point in reference to the master control ey gene sets up for continuation of molecular research in a variety of different directions. The same processes could be done or others, such as DNase footrprinting or probing with the different genes. A separate field experiment could be done with ectopically induced eyes and the survival of those flies, which could utilize many different statistical tests. The possibilities at this point for future research is simply endless! So much knowledge about the underlying reason for diversity in eyes across taxa could be tapped from this point forward.


Campbell, Malcolm. 2002 Introduction to DNA Microarrays. <> 2003 Apr 29.

Michaut, Lydia et al. 2003. Analysis of the eye development pathway in Drosophila using DNA microarrays. PNAS. Vol. 100: 4024-4029.

Weaver, Robert F. Molecular Biology 2nd ed. 2002. New York: McGraw-Hill Companies, Inc. p. 800-802.

If you have questions or comments, please e-mail Maureen Keogh.


Return to Maureen Keogh's Molecular Biology Homepage

Davidson College's Molecular Biology Course Homepage

Davidson College Biology Dept Homepage

Davidson College