The Effects of Relative Fitness of Drosophila Melanogaster on Evolution

Biology 301
12/3/2012
The Effects of Relative Fitness of Drosophila Melanogaster on Evolution

Abstract This experiment was conducted to study the relative fitness of two phenotypes of the Drosophila melanogaster and how fitness can affect evolution in the population. The phenotypes were placed in two different environments, one in which contained a predator and another with no predator. Results of the experiment would show how the fitness of each phenotype is affected by providing a mechanism, and if evolution was occurring in the population. Two hypotheses were inferred, one for each environment. For the cage uninfluenced by a predator, we hypothesized that evolution would occur due to sexual selection, and that sexual selection would be in favor of the wild-type drosophila. For the cage containing the predator, we hypothesized that the vestigial flies would have a higher relative fitness due to natural selection. A ratio of wild-type to vestigial flies was determined, and was set up in each environment. 10 wild-type to 40 vestigial flies was chosen, giving a total of 50 flies for each environment. Each week the flies were fed, and every two weeks they were counted to represent a new generation. At the end of the 13 week experiment, the last generation of flies were counted and recorded in a data table. The results of the experiment show that evolution was occurring in both cages, and that wild type flies were dominant regardless of the environment.
Introduction
The species studied was the Drosophila melanogaster, or more commonly known as the fruit fly. This experiment was conducted to see how the relative fitness of two different phenotypes of the Drosophila melanogaster can affect evolution in the population of two different environments. The relative fitness of an organism is the ability of that organism to both survive and reproduce in its environment. A value of 1 is assigned to an individual is who is best suited for its environment, and all other individuals are assigned a number less than 1 (Kundapur 2008). If a species reaches a relative fitness level of 1, then the population is undergoing fixation, and all other species would be extinct. If no evolution was occurring in the population, then it would be undergoing Hardy- Weinberg equilibrium. The Hardy- Weinberg principle states that both the allele and genotype frequencies remain constant, for each generation, unless disturbances occur in the population. For this transpire, evolution, the change in the inherited characteristics of a population over time, cannot be present (Ko 2007). If evolution is not occurring, then after each generation, the allele frequencies and genotypes of the population would be the same. This means that the ratio of the beginning generation of species must be consistent with the ratio of species to infinite generations.
The two phenotypes of Drosophila used in the experiment were wild-type and vestigial. The wild-type flies, possess the dominant allele, have fully grown wings, and aren’t inhibited to flight. Vestigial flies, which are homozygous recessive, have a mutation to their wings in which they are unable to fly. Males are relatively smaller than female flies and have a small black patch at the end of their abdomen (Santos et al 1997). This is important to know when separating flies, to ensure not all one gender is present in a population, and reproduction is able to occur. Fruit flies have a unique way in which they mate. The male fly, orients its body towards the female and sings for the female fly by spreading its wings and vibrating them rapidly (Kundapur 2008). This would mean if a wild-type fly mates with a female, the offspring will be wild-type, as well as for vestigial.
The experiment would consist of 50 flies in a ratio of 10 wild-type to 40 vestigial flies. Half of each type would be male and the other half female. Two environments would be considered in the experiment, one containing a predator and the other with no predator. The predator used in the habitat was fly paper coated with a sticky substance. The predator would be hung from the top of the cage and wouldn’t touch the bottom, so only the wild-type flies were assumed to be affected. Each habitat would contain 50 flies and the ratio would remain the same. Two hypotheses were given, one for each environment. For the cage lacking predation, we hypothesized that evolution would occur due to sexual selection, and that sexual selection would favor the wild phenotype (McKean et al 2008). This hypothesis was based on the idea of sexual selection, when a female prefers one phenotype over the other. If the vestigial flies aren’t able to do the mating dance, then females are unlikely to be attracted to that specific phenotype (Kundapur 2008). This would give the wild-type flies an advantage, increasing their gene pool from generation to generation, also insinuating that evolution would be occurring.
When considering the cage undergoing predation, we hypothesized that vestigial flies would have a higher relative fitness due to natural selection, and therefore will be undergoing evolution. Natural selection is the process by which organisms have traits that better enable them to adapt to specific environmental pressures (Santos et al 1997). Our hypothesis would make sense considering the vestigial flies, which are unable to fly, are better adapted to their environment, and are out of reach from the predator. This does not change the fact that they may not be as well suited for reproduction, as the mechanism is natural selection and not sexual selection. If natural selection is occurring then evolution must be occurring.
At the beginning of the experiment software called populus was used to measure “p” (dominant allele) allele frequencies over 6 generations. This was done five times, each representing different models. By subtracting p from 1 we are able to find the value of q, giving us the frequencies of both dominant and recessive alleles. The populus data was used as a guide for the final results of the experiment.
The importance of this experiment was to prove how fitness can affect evolution, considering both sexual and natural selection. If evolution was occurring, we know that a mechanism was involved and this could help us understand how and why some species populations are better adapted, and are able to survive longer. This would also help us understand, how over time some genes, are more successful than others and how they give certain organisms an advantage to reproduction. This could also tell us how certain species, which have undergone speciation, are better adapted to their environment than their ancestors, and how their fitness level would be affected if relocated to a different environment when conditions differ.
Material and Methods
To begin the experiment, a ratio of vestigial to wild-type flies had to be made to create a fair comparison. A ratio of 10 wild-type flies and 40 vestigial flies was chosen, giving us a total of 50 fly’s in each cage. Two cages were setup up in which one contained no predator and the other did. The predator introduced in one cage was fly paper that tended to only take out the wild-type species. It was hung from the top of the cage and made sure not to touch the ground. The class was split up into groups and was given a few tubes containing flies. Each row contained either a jar of vestigial or wild-type flies, to keep the experiment organized. Each tube contained a specific type of fly, while also being provided with a sufficient amount of food. The objective was to separate the flies by phenotype and gender to reach our initial ratio of 1:4. This would mean that out of the 10 wild-type flies, half would be females and the other half males, and continue this trend with the vestigial flies. This process had to be repeated twice to have enough flies to support both predator and non-predator cages. To count the flies, fly nap was inserted into one end of the tube and left there until the flies became immobile. The jar was made sure to be kept on its side, so the flies wouldn’t be caught in the food at the bottom of the jar. When deemed immobile, the flies were removed from the container and separated by gender. Male flies are relatively smaller than female flies and are in indication of gender. To be more accurate, male flies contain a black patch at the ends of their abdomen that females do not have, giving a clear indication of the gender of the fly. After the flies were separated, they were put into a new tube which contained food, and handed to the T.A. Once all the flies were counted they were added to the two experimental cages. Each week the flies were fed by mashing up bananas, sprinkling yeast on the food, and putting them into the separate cages. Every two weeks, fly traps were set up by soaking cotton balls in vinegar, and they were counted by species, representing one generation. This was done so for 13 weeks, and then counted by wing type among the class. Each group was given a vile from either the predator or standard cage, and was separated by wing type. The results were recorded and added to our 6th and final generation in the excel spreadsheet. A chi- squared analysis was done for the expected and observed results of the data. The chi squared analysis would tell us if the cage was undergoing evolution and if the data, compared to the populus predictions, were similar. The chi- squared analysis is necessary to support or reject our hypothesis.
Results
What we can determine from the graph is that models 1 and 2 closely represent the data recorded from the class. Model 1 from the populus prediction more closely represent the sexual selection cage as you follow it through generation 6. A huge difference shown is that in generation 2, the p allele frequency drops to .15, a significant amount from generation 1, but almost gets back on track for generations 3 through 6. Model 2 represents the predation cage even though there are no significant similarities between the two. The predation cage data collected shows that in generations 1 and 2 the p allele frequency was very low and that there was a huge increase from generations 3 through 6. Model 2 was only chosen because it ran the most consistent pattern with the predation cage.
Figure 1. Predicted Allele Frequency based on Populus models. Figure 1 shows the predicted allele frequencies of the p allele over 6 successive generations. All models start out with an allele frequency of .2 meaning the q allele is .8 for generation 0. Sexual and predation data was added to represent the statistics from the actual experiment in comparison to the populus predictions.
Figure 1. Predicted Allele Frequency based on Populus models. Figure 1 shows the predicted allele frequencies of the p allele over 6 successive generations. All models start out with an allele frequency of .2 meaning the q allele is .8 for generation 0. Sexual and predation data was added to represent the statistics from the actual experiment in comparison to the populus predictions. Table 1 represents the significant differences between the observed and expected patterns of Hardy-Weinberg Equilibrium and the populus model predictions. Both environments show a significantly low p-value for the Hardy-Weinberg equilibrium comparison. In the cage lacking predation, we see for model 1, the p-value is larger than our alpha of .05, showing that there is no significant difference. For model 2, in the predation cage, we again see a significantly low p-value. From table 1, we are able to support or reject our hypothesis, and determine if evolution was occurring in the Hardy-Weinberg row of the data.

Column1 | Sexual/ Natural Selection Only | Predation | Hardy- Weinberg | 4.8091E-83 | 2.3308E-166 | Model 1-2 | 0.29436543 | 6.37959E-72 |

Table 1. P-Values of Expected to Observed Phenotypes. The chi squared analysis shows the p- values taken from the expected and observed numbers of vestigial and wild-type flies. The table shows the p-values related to our hypotheses. The p- values recorded were for models 1 and 2 in figure 1 above, and for data recorded compared to Hardy-Weinberg equilibrium.
Table 1. P-Values of Expected to Observed Phenotypes. The chi squared analysis shows the p- values taken from the expected and observed numbers of vestigial and wild-type flies. The table shows the p-values related to our hypotheses. The p- values recorded were for models 1 and 2 in figure 1 above, and for data recorded compared to Hardy-Weinberg equilibrium.

Discussion The data from figure 1 gives us a general idea of which models the sexual selection and predation cages are most closely related to. Cage 1 with no predation is most closely related to model 1 as the graphs look almost identical except for the spike in generation 2. When looking at the sexual selection cage, we see that part of our hypothesis was supported. Wee hypothesized that the wild-type flies would be the favored phenotype due to sexual selection (McKean 2008). The p-allele (dominant) seemed to be more successful than the q-allele (recessive) through 4 of the 6 generations. Generations 2 and 6 show a drop in wild-type allele frequency. This could be due to the fact that no predator is affecting the wild phenotype population, and the males are more reproductively fit than the vestigial flies, as they are able to do the mating dance more efficiently. Frequent changes in the data show that evolution is occurring as the frequencies of the genotypes fluctuate and do not remain the same (Ko 2007). Table 1, confirms that the p-value between the expected and observed data from the class set was extremely low, having a value of 4.8091E-83, meaning that we reject the null hypothesis that cage 1 was under Hardy-Weinberg equilibrium. This number is extremely low and shows a significant difference between the observed and expected data, emphasizing that evolution has occurred, further supporting our hypothesis that evolution would occur due to sexual selection. Also, table 1 shows that the p-value of model 1 is relatively high compared to our alpha. A p-value of 0.29436543 was given, showing us that there was no significant difference between expected and observed data in the sexual selection cage. Since a higher p-value is given, we reject the null hypothesis, as we want the data to be similar. This also supports our hypothesis that the wild-type fly would remain dominant and would do so because of sexual selection. Knowing that both parts of our hypothesis were supported, we are able to conclude that sexual selection was a major mechanism of evolution. The female flies preferred mating with the wild-type phenotype as they were able to do the mating dance successfully. This would mean that wild-type fly, in the cage with no predation, had a higher relative fitness. Our second hypothesis was that the vestigial phenotype would have a higher fitness level due to natural selection. When looking at table 1, we know that evolution is occurring because the p-value of 2.3308E-166 in the predation cage is significantly lower than our alpha. This means we reject the null hypothesis that no evolution was occurring. Since a predator is introduced in this cage, and it was assumed that its tendency was only to affect the wild-type population, the mechanism of evolution must be natural selection, since the vestigial flies are better adapted the this particular environment (Santos et al 1997). Figure 1, shows that in the predation cage, the vestigial flies were dominant in 5 of the 6 generations. Their dominance was clear in early generations, but as time went on, the wild phenotype relative fitness increased significantly, until generation 6 when they had a higher fitness level than the vestigial phenotype. Although the vestigial phenotype remained dominant for most of the successive generation, the data does not strengthen our hypothesis as the wild phenotype fitness level was able to increase at a significant rate. Table 1 also shows that for model 2 there was a significant difference between our observed and expected data as the p- value was 6.37959E-72. This concludes that we fail to reject the null hypothesis, and that our hypothesis is weakened from the results. However, some predictions as to why these events occurred were assumed. From generation to generation, the predator in the cage remained in the same spot and did not move. This could mean that the wild phenotype Drosophila could adapt to the location of the predator, thus avoiding it (Priest et al 2008). If this is true, than sexual selection would reoccur, as the wild-type flies would continue to be more efficient at reproduction. Both cages proved that Hardy-Weinberg equilibrium was not present as both cages had a mechanism of evolution involved. This means that both cages were undergoing frequent changes in allele frequencies and genotypes, supporting both our hypotheses. To obtain better results in future experiments, it would be more suitable to run the data over more than 6 successive generations, to receive more accurate data over time. Also, if the predator were moved to different areas of the cage, we would be able to better see if the vestigial flies are better adapted to a predation environment. On an ecological scale, the results of the experiment show that fitness level is directly related to the environment. Although overtime, organisms adapt to survive, they are more inclined in early generations to have a lower relative fitness (Mery 2007). These sets of data, on a larger scale, could show us how we as humans, over time are able to maintain a relatively high fitness level as our environment changes. When obstacles are put in our way, we are eventually able to find a solution and overcome them. For example, if someone has the flu, he or she must treat the illness with the necessary steps. In the future, more precautions would be taken to ensure that such an illness would not occur again. If under Hardy- Weinberg equilibrium, we would not be able to adapt to our environment, and the genotype and illnesses we possess now, would remain indefinitely.
Acknowledgments
I would like to thank the University of South Carolina Biological Sciences department for providing me with the materials and space to perform this experiment. I would also like to thank Katy Hallman, and Taylor Smith for assisting in the overall experiment. Each member of the group was able to perform each part of the experiment, and was done so equally.
Reference
Kundapur J. 2008. The evolution of fitness after prolonged sperm storage in drosophila melanogaster: Adaptation by both sexes.
Mery F. 2007. Experimental evolution of learning ability in drosophila melanogaster: Benefits and fitness costs of learning. 2007 Conference of the Society for Experimental Biology (SEB 2007) .
Priest NK, Galloway LF, Roach DA. 2008. Mating frequency and inclusive fitness in drosophila melanogaster. Am Nat 171(1):10-21.
Ko W. 2007. Molecular phylogeny and evolution in the drosophila melanogaster species subgroup.
McKean KA, Nunney L, Hughes K. 2008. Sexual selection and immune function in drosophila melanogaster. Evolution 62(2):386-400.
Santos M, Borash DJ, Joshi A, Bounlutay N, Mueller LD. 1997. Density-dependent natural selection in drosophila: Evolution of growth rate and body size. Evolution 51(2):420-32.