Skip to main content MCC Crescent Shape Search rss Facebook Twitter Youtube Instagram chevron pointing down

Fly Eye - Red Pigment


Fly eye color variation showing different amounts of red pigment and brown pigment. Wild type (brick red) is shown on the upper right with both brown and red pigments.  Cinnabar (upper left) shows no brown pigments.  Sepia (bottom) shows dark brown with yellow pigment. The fly in the center is homozygous for both sepia and cinnabar, showing light brownish yellow eyes.

Fly Eye

Examination of Red Pigment Production in Drosophila


The wild type (brick red) eye color in Drosophila is due to a mix of brown pigments (ommochromes) and bright red pigments (pteridines, especially drosopterin).  Drosopterin is orange under ultraviolet light, while other pteridines are blue or yellow. For purposes of this project, “pteridines” will be used to refer to all the related pigments of the red pigment pathway, whether red or not. When discussing individual pigments, the color of the pigment will usually refer to its appearance under ultraviolet light, even though, for example, “orange” (drosopterin) makes the wild type eye red.  The two pigment pathways are very different chemically and synthesized in different pathways. Numerous mutations have been found in both pathways. A few mutations affect both pigment pathways, such as the sex-linked white eye locus. It is possible that other mutations affect both pathways. Often similar phenotypes are caused by different mutations. The red pigment pathway intermediates (pteridines) have a well worked out protocol for chromatography and will be emphasized in this study. However, being able to block brown pigment production allows mutations of red pigment production to be viewed more easily without the masking effect of the brown pigment. Essentially this is what is also observed in the chromatography paper since the brown pigments are not dissolved in the solution and moved up the paper. In fact, one of the main goals of this research project is to be able to compare the chromatogram of red pigment mutations with the actual red pigments in the intact eye without the masking effect of brown pigment. The other main goal is to determine the position of red pigment intermediates and color of the pigments produced at each step.

In order to view the steps at which red pigment production is stopped by a particular mutation, it is necessary to block the masking effects of the brown pigment. Three mutations have this effect and will be used: scarlet, vermilion, and cinnabar. Which one of these is used depends on the location of the red pigment gene under investigation, so that the genes will independently assort to give dihybrid ratios. Most red pigment genes are on chromosomes II and III. Vermilion is on the X (or number I) chromosome and can be crossed with most red pigment mutations but adds the complication of modified 9:3:3:1 ratios due to sex-linkage. Scarlet is on chromosome III and cinnabar on chromosome II. One last note to the student, please try to keep the names of mutations clear. A gene is named by the phenotype of the mutation. Since mutations of red pigment production reduce red pigment, they will show predominately brown pigment, hence the names brown, purple, rosy, and sepia refer to red pigment mutations.

In order to determine the order of the steps, the red eye pigment genes are crossed with each other. The earlier step in the sequence will be epistatic to a later step since the later step will have no precursor to act on. Think of an assembly line here.  If the worker first in line fails to do their job, then the result will be the same whether the second worker in line was capable of doing their job or not.  If a mutation blocks a particular step, then the pigments that are visible in the mutant are actually the ones produced prior to that step. For example, if brown eye shows no red pigments and purple shows yellow and blue pigments, then the brown gene (or specifically the wild type of brown) was necessary for the production of yellow and blue. Also, keep in mind that pigment pathways may not be strictly linear but can be branched so that a block at one step may cause a build-up of pigments of several previous steps, including some that branched off in more than one direction.

Since large numbers of flies are required (and fly counters!), any one group of students will be limited to working on one step. For example, we want to know what the wild version of the purple gene and the wild type of the sepia gene each do. If we cross them, we should be able to determine which comes first in the pathway. We can speculate that the pigments seen in the second mutation are those that are produced by the wild type of the first mutation in the pathway. However, since there might be a step in between the two genes examined in one cross, that step will remain for a future cross to determine.

On the other hand, if we are not interested in the order of the steps, but in the actual appearance of the pigments present in a mutant (not what the wild version of the gene could make but what the mutant accumulates), then we cross it with an independently assorting brown pigment blocker. So for example, we want to see what red pigments are in the purple eye, so we cross it with scarlet and get some F2 that are homozygous  for both purple and scarlet showing the red pigments in the purple eye without the masking brown. We can view these in the intact eye under visible (white) light and compare those to the chromatogram of purple eyes under ultraviolet light. Keep in mind that pigments under UV light radiate visible light that is somewhat different than what they reflect under white light.

Let’s get started. Let’s choose a mutation from column A. If our goal is to find its position in the sequence of red pigment production, then choose another mutation from column A. If our goal is to see what pigments are present in the mutation, then you should use the appropriate (independently assorting) mutation from column B. (Brown will be used to find the position of column A mutations only  and not for pigmentation since it produces no pteridines.)

Column A Chromosome Column B Chromosome
brown II scarlet III
purple II vermilion X
sepia III cinnabar II
rosy III    

Use of symbols for alleles. This is not a technical website, but is specifically for use by beginning students who frequently make Punnet squares by hand. Hence, easy to distinguish upper and lower case, one-letter symbols are used.  All of the mutations here are recessive and use the lower case and wild type has the upper case. The following are used: b = brown, p = purple, d = sepia, r = rosy, e = scarlet, n = cinnabar, and Xl = vermilion (sex-linked).


Fly crosses

Fly stocks were obtained from Carolina Biological Supply except where otherwise noted. Fly crosses are done by standard procedures by mating virgin females of one type with males of the other type. Reciprocal crosses are done. Parents are cleared after one week to prevent breeding with the next generation. First generation offspring (F1) are placed in new media with approximately 10-15 flies of each sex. The second generation offspring (F2) are counted after two weeks (about the fifth week of the entire experiment).  Beginning with two mutant types and wild type, there will be six total crosses started for any given experiment.  Students are trained to recognize eye colors by examining examples placed side-by-side by the instructor.  The instructor will be available for help in determining eye colors but will encourage students to work confidently, examining each fly at about 15X total magnification, moving them into separate piles, but not deliberating over individual flies. The instructor will then look at the separated piles to make sure that students apply the proper label to each group but generally not over-ruling the student if only a few flies appear to be misplaced. My experience is that students working confidently make fewer mistakes than those that deliberate too long. An individual student may have a consistent bias toward assigning questionable flies to a certain group, but since other students often have a different bias, the biases will tend to cancel out. Of course large numbers of flies with several different students counting is the best situation. The probability of the deviation of observed numbers from expected numbers of the stated null hypothesis is calculated by CHISQ.TEST and given for each cross.


 Carolina Biological Supply materials are used, and their method of chromatography is run by students. The system is modified from that of Hadorn and Mitchell (1951) because of faster movement of pigments. Crushed fly heads are placed about ½ inch from the bottom of the chromatography paper, just above the level of the solvent, and run for about 90 minutes in foil-covered jars. The papers are then removed and air-dried before viewing with a middle wavelength UV transilluminator (Edvotek #558). Two males and two females of one type are run on the same paper as two males and two females of the other type. All combinations of three types of flies two at a time are done in any given run so all side-by-side comparisons can be made.  


Results are given for pteridine pathway mutations. The positions of the mutations in the pathway and the pigments produced by the mutations are compared with each other.

 See results for purple.

See results for sepia.

See results for rosy.

Chromatograms of the mutant phenotypes for red eye pigments examined in this study, including sepia, purple, rosy, and sepia-rosy. Brown (ommochrome) pigments do not show up in chromatograms. From left to right, the chromatogram shows sepia female, purple male, rosy female, rosy male, and sepia-rosy female.



Because of small sample sizes for most crosses, definite conclusions can be hard to draw. It is difficult, for example, to determine what pigments are produced at each step.  Since purple mutants (obtained from the Bloomington stock center) have blue, yellow, and orange pigments, the defective enzyme must reduce, but not eliminate, production of yellow and orange. Blue pigments are present in all chromatograms so must be produced prior to the steps catalyzed by wild type for purple (see results for purple Figure 1 and Figure 2) and wild type for sepia.  Because of apparent epistasis of purple over sepia as shown in Table 3, noting particularly the total for crosses 3 and 4, the step catalyzed by the product of wild type for purple must occur at a position before the step catalyzed by wild type for sepia. This is consistent with the red pigment pathway summarized by Kim et al. (2013).

The sepia chromatogram is puzzling.  Yellow, as expected for this eye color, is clearly visible under white light (not shown) on the chromatogram. However, under ultraviolet it is not seen, but blue and blue-green are very distinct (see Results for sepia Figure 1 and Figure 2). Orange is not seen, so it is apparent that the step catalyzed by wild type of sepia is necessary for orange pigment production. Either the large amount of blue and blue-green obscures yellow under ultraviolet light or yellow is altered in some way. Reaume et al. (1991) noted that sepiapterin broke down quickly in UV light to give another unspecified pigment. According to Nassau (1983), yellow transmission is the result of absorption of violet. If the concentration of yellow pigment is high enough, absorption will also be significant in blue and blue-green, and it will appear orange or even brown. It seems possible that if the high concentration also absorbs into the UV range, which is intensified under a UV lamp, then blue and blue-green will be emitted (fluoresced), possibly obscuring the transmitted yellow pigment.

In the intact eye, double mutations for purple or sepia along with a brown pigment blocking mutation (cinnabar, scarlet, or vermilion) give a yellowish color to the intact eye. Blue pigment should be present but is not visible under white light. Once again, purple produces at least some yellow pigment and sepia produces even more, as clearly seen under white light in the intact eye. In the case of sepia the eye darkens with age so that the double mutant begins to resemble the dark brown of the sepia fly (compare Figure 2 with Figure 3 in Results for sepia). Somehow, the pteridines present in sepia produce a brown color, perhaps due to the concentrating effect mentioned above (Nassau, 1983), even in the absence of ommochrome brown pigments.

The ommochrome brown pigment production steps catalyzed by wild type for scarlet, cinnabar, and vermilion all seem to be necessary for any brown ommochrome pigment production and my also reduce red pigment, at least for a time. As such, these mutations are interchangeable with regard to blocking brown to view red pteridine pigments in the intact eye. However, they may not be interchangeable in regards to viability of flies homozygous for one of these mutations along with a pteridine pigment mutation. Cinnabar appears to have a much greater effect in reducing viability although all double homozygotes were observed in less than expected numbers (see Results for sepia Table 1, crosses 5 and 6).  Of course, it must be kept in mind that these pigment chemicals are used in other metabolic pathways than just producing eye color, for example, as coenzymes. Another view of pteridine synthesis is given in Kamleh et al. (2009). Thin layer chromatography analyses for pteridine pigments are given in Reaume et al. (1991), with particular emphasis on the rosy locus.

The rosy locus was the third pteridine pigment pathway mutation examined in this project. According to pathways given by Kim et al. (2013) and Kamleh et al. (2009), rosy wild type is not involved in the direct pathway to drosopterin, the most important red pigment in the wild type eye. According to Reaume et al. (1991) the wild type for rosy is the structural gene for xanthine dehydrogenase. The enzyme catalyzes the conversion of 2-amino-4-hydroxypteridine to isoxanthopterin (Mertens and Hammersmith, 2007;  Reaume et al., 1991). Rosy eyes should lack blue-violet pigment(isoxanthopterin) although this does not affect the appearance under white light. Rosy was crossed to cinnabar and vermilion to help determine what pigments are lacking due to rosy and, hence, what is produced by the wild type allele. Rosy also seems to accumulate sepiapterin (yellow) as in sepia discussed above, as it also darkens (turns brown), beginning to resemble sepia; rosy also resembles wild type with age, probably due to accumulation of drosopterin.

The rosy mutant was crossed with purple to help position the sequence of these two loci in the reaction sequence or network because both involve pteridine pigment production steps. With regard to position of rosy (see Results for rosy Table 1 cross 6), rosy appears to be epistatic over purple, even though rosy depends on the product of purple (pyruvyl tetrahydrobioptern), which is converted into xanthopterin, in turn converted to isoxanthopterin by rosy wild type (Kamleh et al., 2009). Since the purple mutant allows some production, the excess product not converted into isoxanthopterin must build up and be converted into sepiapterin, making the eye darker than purple.

One last cross was performed to see the interaction of sepia and rosy, both of which are on chromosome III and, hence, linked and not independently assorted. The sepia-rosy stock was obtained from the Bloomington stock center (stock #37751). Unfortunately, little data were obtained (see end of sepia results). The double mutant eyes appeared very similar to sepia although they may have been slightly darker. Since linkage maps indicate about 25% recombination, I incorporated that into the null hypothesis. However, the hypothesis was rejected for both pooling of sepia-rosy with sepia and for keeping them separate. The chromatograms were also similar as can be seen above.

Red Pigment Production Summary 

Literature Cited

Hadorn, E. and H.K. Mitchell. 1951. Properties of mutants of Drosophila melanogaster  and changes during development as revealed by paper chromatography. Proc. Natl. Acad. Sci. USA 37: 650-665.

Kamleh, M.A., Y. Hobani, J.A.T. Dow, L. Zheng, and D.G. Watson. 2009. Towards a platform for the metabonomic profiling of different strains of Drosophila melanogaster using liquid chromatography-Fourier transform mass spectrometry. FEBS Journal 276: 6798-6809.

Kim, H., K. Kim, and J. Yim. 2013. Biosynthesis of drosopterins, the red eye pigments of Drosophila melanogaster. IUBMB Life 65: 334-340.

Mertens, T.R. and R.L. Hammersmith. 2007. Genetics laboratory investigations. Pearson-Prentice Hall. Upper Saddle River, pp. 211-217.

Nassau, K. 1983. The physics and chemistry of color. John Wiley & Sons, Inc.pp. 111, 287. 

Reaume, A.G., D.A. Knecht, and A. Chovnick. 1991. The rosy locus in Drosophila melanogaster : xanthine dehydrogenase and eye pigments. Genetics 129: 1099-1109.

Ziegler, I. 2003. The pteridine pathway in zebrafish: regulation and specification during the determination of neural crest cell-fate. Pigment Cell Research 16:172-182.

Dr. Stephen Reinbold

Last Modified: 6/1/18