Printer Friendly

When a fly has to fly to reproduce: selection against conditional recessive lethals in Drosophila.

[ILLUSTRATION OMITTED]

Teachers of evolutionary biology at the high school and college levels often need hands-on, inquiry-based laboratory exercises that allow students to observe and explain fundamental evolutionary processes. In its National Science Education Standards, the National Research Council has emphasized the fundamental importance of inquiry-based teaching approaches that incorporate real-life experiments (Olson and Loucks-Horsley, 2000). In many situations, these are preferable to traditional instruction or use of simulations. In evolutionary biology, a strong obstacle to developing such educational activities is the difficulty of observing microevolution in action within a semester. Both available technology and time constraints attract more and more college and high school teachers to various simulation activities, computer-based or otherwise. While some of these activities are attractive and efficient (e.g., Dubowsky & Hartman, 1986; Soderberg & Price, 2003; Rodriguez & Dvorsky 2006), a live experimental design has unique value. A number of such exercises (both field and laboratory) have been described in the recent literature (McComas, 1991; Hilbish & Goodwin, 1994; Salata, 2002; Vondrasek et al., 2004; Coleman & Jensen, 2007), yet there is a shortage of educational experimental designs that allow students to observe natural selection in action. One of the reasons why it has been difficult to bring microevolutionary experiments into classrooms is the obvious tradeoff between choosing an allele that is strongly selected against, such as a lethal allele, and choosing one that is easy to visualize. Alleles in the latter category are often subject to weaker selection, which means that a change in allele frequency takes longer to detect. For example, observing selection against a visible marker in Drosophila in an educational setting can be quite straightforward (Salata, 2002), but most visible mutations in Drosophila experience only weak stabilizing selection (e.g., Yampolsky et al., 2005), so it may take more generations to observe a significant change in allele frequency than is possible within a teaching-lab framework. Selection against lethal mutations--or sterility mutations, which, from the evolutionary standpoint, are identical--is strongest and therefore fastest. But both lethal mutations, most of which act early in embryonic development, and sterility mutations are hard to observe phenotypically. A good compromise in addressing this dilemma is to use conditional mutations, that is, mutations that cause death (or sterility) under some conditions but are viable (fertile) under other conditions.

In designing such experiments, it is most straightforward to use microorganisms, given their short generation time and readily available conditional lethal mutations. For example, auxotroph mutants are easy to maintain and observe on selective media and are strongly selected against on restrictive media (Welden & Hossler, 2003; Krist & Showsh, 2007). However, a disadvantage of using microorganisms for this purpose is that technical details of handling and propagating the cultures may obscure the evolutionarily essential components of the experiment. Thus, it is desirable to develop a selection experiment that uses a more familiar, eukaryotic diploid organism that offers the advantages of conditional lethal mutations. Drosophila, which has been the mainstay model organism in genetics for almost a century, is a quickly developing, easy-to-culture organism with a generation time of as little as 9 days. Culturing and handling require no special equipment, and stocks, culture components, and laboratory manuals are readily available (Ashburner et al., 2005; Carolina Biological Supply Company, 2008). Auxotrophic mutants are available in Drosophila (e.g., ade2; Johnstone et al., 1985), but they are hard to visualize and require special minimal media for selection to operate. We suggest a simple experimental design in which natural selection acts against a phenotypically visible conditional recessive lethal, thus allowing observation of significant allele frequency changes in just a few generations. Specifically, we propose using selection against flightless mutants such as apterous (ap) or vestigial (vg) in population cages in which the food source used to establish each new generation is accessible by flight only. To the best of our knowledge, this is the only published laboratory experiment using Drosophila--and possibly the only one using any metazoan--that is suitable for observing natural selection against recessive lethals within one semester-long exercise.

The proposed activity is methodologically straightforward and possesses several pedagogical advantages that make it particularly suitable for a high school instructor aiming to foster an inquiry-based interdisciplinary approach. Discussion of the frequencies of mutant and wild-type phenotypes in each generation presents a good context for introducing the ideas of stochasticity and probability in life sciences. The data that students collect during this experiment are suitable for a straightforward statistical analysis and provide a context discussing the use of graphs in representing scientific results. Furthermore, this hands-on illustration of the mechanism of natural selection is a good example of the epistemological shift from "knowing that" to "knowing how," which is often lacking in science curricula.

Materials & Methods

We used Carolina Biological D. melanogaster stocks ap-se, homozygous for the ap allele and an unlinked sepia (se) allele, and vg-e, homozygous for the vg allele and an unlinked ebony (e) allele (Carolina Biological ER-17-2753 and ER-17-2760, respectively). Flight-unrelated visible markers serve as the control for any flight-unrelated selection pressure against genotypes brought into the population from these stocks. The wild-type component of the selection populations was D. melanogaster lines recently extracted by the authors from a natural population in east Tennessee.

Eight population cages were set up using Sterilite[R] white 20 qt/19 L plastic containers with mesh windows cut in the lids. To eliminate the linkage disequilibrium between the flightless allele and the visible marker, each population was started by placing into each box an open bottle containing 100 flies heterozygous for either ap and se alleles or vg and e alleles (4 boxes each; there is no need to collect unmated flies, because all the flies are heterozygous and all matings in this and consecutive generations are random). Additionally, another bottle containing standard Drosophila medium was suspended from the lid on a nylon string (Figure 1). The string and the edges of the lid were lined with 3M[R] silicon grease to prevent flies from walking into the food. This suspended food bottle was used to establish the new generation in each box. Generation time was 2 weeks. We did not take total population census each generation, but the bottles typically contained approximately 1000 pupae at the time of transfer into the fresh box.

In each generation, when the food bottle was transferred into the new box, adult flies were anesthetized by CO2 (placing the population box in a fridge for 10 minutes works as well) and a sample of adults flies (200-400 from each box) was taken and phenotyped into four categories: wild type, wingless, ebony-bodied/sepia-eyed, or doublemutant. Each box can be assigned to a student or a pair of students, and handling and counting the flies takes about 2 hours for each generation. We continued the selection experiment for seven generations (3.5 months).

Because mating in the population boxes occurs both on the new (suspended) food source and on the old food source on the bottom of the box, selection against wingless alleles would not be equivalent to selection against recessive lethals if wingless males could mate with winged females. To test that possibility, two population cages (one with each wingless mutant line) were set in the same manner as the regular cages in the selection experiment, except that the population was made up of 30 homozygous mutant males, 120 wild-type males, 30 homozygous mutant females, and 120 heterozygous females. Thus, this experiment simulates the situation observed in the first generation of selection with frequency of mutant phenotype of 25%. After 4 days of mating, the flies were removed from the boxes, wild-type females were placed individually into small shell vials containing 1 ml of food, and the phenotypes of 20-40 offspring that emerged in these vials were scored. Any mutant offspring would indicate at least some mating between wingless males and winged females. No such mutant offspring were observed, which excludes the possibility that the portion of winged females that mated with wingless males at least once exceeds 0.035 with significance lever p=0.05 (Zar, 1999, p, 527). Thus, there is little chance that homozygous males will produce offspring in this experiment, and any deviations from predictions based on selection against recessive lethals are quite small and probably negligible given the typical sample sizes.

[FIGURE 1 OMITTED]

Allele frequencies at each of the four loci were calculated with the assumption of Hardy-Weinberg equilibrium. The expected frequency of flightless mutants was estimated according to discrete generation selection against a lethal allele (Crow & Kimura, 1970: eq. 5.1.1):

[q.sub.t] = [q.sub.0]/1 + [q.sub.0] t (1)

where [q.sub.t] is the frequency of the allele in generation t and [q.sub.o] is the initial frequency of the allele (i.e., 0.5 in our experiments). Experimental points located below this expected curve would indicate partial dominance (i.e., recessive flightless mutations reduce fitness in heterozygous state). Data points located above the expected curve would indicate either that lethality is not complete (some flightless flies manage to reach the suspended food) or that there is some selection in favor of heterozygotes. Thus, this experiment invites useful discussion of a variety of data analysis and interpretation issues.

The frequency of flight-unrelated visible markers is not expected to change drastically during the experiment, because they are not linked to the flightless mutants and both e and se mutants usually have higher fitness than the wild type. As expected, no phenotypic linkage disequilibrium between flightless phenotypes and marker phenotypes was observed, with the exception of 2 out of 4 replicate populations in generation 1, in which ap and se phenotypes were at a significant phenotypic linkage disequilibrium (chi-square test, p < 0.0015 and p < 0.003). On the other hand, a decrease in visible allele frequency can be interpreted as selection against a recessive deleterious mutation. Data analysis and interpretation are less straightforward in this case than in selection against recessive lethals, because there is no analytical solution for qt similar to Equation 1. Thus, estimation of the selection coefficient from the data, although possible (see Appendix), is probably of little didactic value. However, Equation 3 (Appendix) can be used to numerically plot curves describing selection against nonlethal harmful mutations to emphasize the similar but slower operation of selection (see below).

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Results & Discussion

As expected, the frequency of both flightless mutants dropped in all replicate populations (Figure 2). Both mutants, however, exhibited a slight deviation from the expected curve toward less intense selection that was, in some generations, statistically significant. The two flight-independent markers differed from each other and from flightless alleles in their behavior: the e allele was selected against, whereas frequency of the se allele did not change during the experiment (Figure 3). Thus, the proposed experimental system allows observation of the action of natural selection on allele frequencies in Drosophila within a relatively short laboratory exercise and provides data that can be used in class to discuss several quantitative evolutionary processes.

That selection against flightless mutants was less intense than the theoretical prediction is unexpected. While the opposite direction of deviation can be readily explained as selection against heterozygotes, the pattern we observe can either be explained by flightless flies somehow gaining access to the suspended food bottle or by selection in favor of flightless alleles in the heterozygous state. Our data did not allow us to test these hypotheses. However, for educational purposes, the exact explanation may be less important than the question's value as a topic for discussion.

Other topics for discussion include possible sources of errors in data analysis. For example, the calculation of allele frequencies relies on the assumption of Hardy-Weinberg equilibrium, possible deviations from which include nonrandom mating due to the separation of winged and wingless flies (if mating tends to occur on fresh food accesible to winged flies only). Another possibility for deviation from Hardy Weinberg equilibrium occurs in flies that are homozygous for the e allele, since such flies have abnormal circadian rhythms of locomotion (Newby & Jackson, 1991), which may result in a rhythm of diurnal mating behavior different from that in the wild types. Last but not least, a teacher can point to situations in which the direction of selection is reversed and flightless genotypes are favored, as in flightless insects on oceanic islands (unknown in Drosophila, but common in other dipterans; e.g., Hardy & Kohn, 1964), thus illustrating the effects of variable selection.

Finally, it should be noted that the use of conditional lethals, while allowing observation of a microevolutionary change within a manageable period, may also unintentionally create or reinforce a belief that selection primarily occurs in gross mutants and in short time frames. It is important to make sure that students understand that selection normally operates on small variants over significant spans of time. One way to accomplish this may be to use Equation 2 (Appendix) to plot allele frequencies during selection acting on a recessive mutation with selection coefficient s<<1, which will allow students to observe that such selection is very slow.

DOI: 10.1525/abt.2010.72.1.4

APPENDIX. Estimation of selection coefficient from data on allele frequency in discrete generations.

The change of recessive allele frequency as the function of this frequency in the previous generation q and the selection coefficient s is calculated as (Crow & Kimura, 1970: eq. 5.2.18; assuming fully recessive allele)

[DELTA]q = -s(1 - q)[q.sup.2]/1 - [sq.sup.2] (2)

which can be rewritten as the estimate of the selection coefficient s for each generation:

s = [DELTA]q/[q.sup.2]([DELTA]q - 1 + q) (3)

Using Equation 3, the selection coefficient can be estimated for each generation separately and the obtained values averaged. This approximate estimate of s tends to have high standard errors because of random variation in allele frequencies and sampling error.

Acknowledgments

We are grateful to Karl Joplin, Foster Levy, and anonymous reviewers for useful suggestions at various stages of the preparation of the manuscript and to students in Evolution class for participating in the experimental lab. LY was supported by NSF-0525447.

References

Ashburner, M., Golic, K.G. & Hawley, R.S. (2005). Drosophila: A Laboratory Handbook, 2nd Ed. cold Spring harbor, NY: cold Spring harbor Laboratory Press.

Carolina Biological Supply company. (2008). Carolina Drosophila Manual. Burlington, Nc: carolina biological Supply company.

Coleman, S.W. & Jensen, J.S. (2007). male mating success: preference or prowess? Investigating sexual selection in the laboratory using Drosophila melano gaster. American Biology Teacher, 69, 351-358.

Crow, J.F. & Kimura, M. (1970). An Introduction to Population Genetics Theory. Minneapolis, MN: Burgess.

Dubowsky, N. & Hartman, E.M., Jr. (1986). Simulated laboratory/field study of evolution. Journal of College Science Teaching, 15, 464-465.

Hardy, D.E. & Kohn, M.A. (1964). Insects of Hawaii, vol. 11: Diptera: Brachycera, Family Dolichopodidae.

Hilbish, T.J. & Goodwin, M. (1994). A simple demonstration of natural selection in the wild using the common dandelion. American Biology Teacher, 56, 286-290.

Johnstone, M.E., Nash, D. & Naguib, F.N.M. (1985). Three purine auxotrophic loci on the second chromosome of Drosophila melanogaster. Biochemical Genetics, 23, 539-555.

Krist, A.c. & Showsh, S.A. (2007). Experimental evolution of antibiotic resistance in bacteria. American Biology Teacher, 69, 94-97.

Mccomas, W. F. (1991). Resources for teaching evolutionary biology labs: an analysis. American Biology Teacher, 53, 205-209.

Newby, L.M. & Jackson, F.R. (1991). Drosophila ebony mutants have altered circadian activity rhythms but normal eclosion rhythms. Journal of Neurogenetics, 7, 85-101.

Olson, S. & Loucks-Horsley, S., Eds. (2000). Inquiry and the National Science Education Standards: A Guide for Teaching and Learning. Washington, DC National Academy Press.

Rodriguez, J., Ortiz, I. & Dvorsky, E. (2006). Introducing evolution using online activities in a nonmajor biology course. Journal of College Science Teaching, 35, 31-35.

Salata, M. (2002). Evolution lab with Drosophila. Bioscene, 28, 3-6.

Soderberg, P. & Price, F. (2003). An examination of problem-based teaching and learning in population genetics and evolution using EVOLVE, a computer simulation. International Journal of Science Education, 25, 35-55.

Vondrasek, J.R., Antonovics, J. & Taylor, D.R. (2004). Evolution kills: a Web resource for instructors of evolutionary biology. Bioscene, 30, 3-8.

Welden, c.W. & Hossler, R.A. (2003). Evolution in the lab: biocide resistance in E. coli. American Biology Teacher, 65, 56-61.

Yampolsky, L.Y., Allen, c., Shabalina, S.A. & Kondrashov, A.S. (2005). Persistence time of loss-of-function mutations at nonessential loci affecting eye color in Drosophila melanogaster. Genetics, 171, 2133-2138.

Zar, J.H. 1999. Biostatistical Analysis, 4th Ed. Upper Saddle River, NJ: Prentice Hall.

ANDREA D. PLUNKETT was an undergraduate student in the Department of Biological Sciences at East Tennessee State University when this article was written, and she is now an M.S. student at the University of cincinnati. LEV Y. YAMPOLSKY is Associate Professor of Biological Sciences at East Tennessee State University Johnson city, TN 37614; e-mail: yampolsk@etsu.edu.
COPYRIGHT 2010 National Association of Biology Teachers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Plunkett, Andrea D.; Yampolsky, Lev Y.
Publication:The American Biology Teacher
Article Type:Report
Geographic Code:1USA
Date:Jan 1, 2010
Words:2809
Previous Article:Encouraging new manuscripts in critical areas of biology education.
Next Article:Using a 17th-century experiment as a gateway to critical assessment of scientific reports.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters