This structured set of lab activities allows students to explore the evolution of pelvic spine reduction in stickleback fish. The exercise draws upon the field of evolutionary and developmental biology (evo-devo) and information presented in the HHMI Holiday Lecture entitled “Fossils, Genes, and Embryos.” Students analyze fossil data from a rich stickleback deposit in Nevada, documenting the evolution of pelvic spine reduction in a preserved population, and then use Hardy-Weinberg analysis to explore the role of natural selection in this type of evolutionary event. Finally, students use molecular genetics and polymerase chain reaction to uncover the evolutionary role of gene switches in pelvic spine reduction. Collectively, the lab activities explore a specific evolutionary event from the combined perspectives of fossil evidence, natural selection, and molecular genetics. The lab also serves as a good introduction to the concepts of gene switches and evo-devo.

Introduction

Evolution is the unifying principle in biology because it explains the overwhelming diversity of form and function of life on Earth. Fossils, biogeography, comparative anatomy, embryology, and genetics all provide evidence for evolution. Nonetheless, evolution can be a challenging topic to teach in the laboratory, in that it cannot be observed directly. As such, lab activities based on actual scientific research are an invaluable resource. In the following set of laboratory exercises, students use scientific data to investigate and explore a specific evolutionary event from the combined perspectives of fossils, population genetics, and molecular genetics. The activities address Next Generation Science Standards HS-LS3 and HS-LS4 and are suitable for an undergraduate biology course or an Advanced Placement high school course. Specifically, students investigate the evolution of pelvic spine reduction in a fish, the threespine stickleback (Gasterosteus aculeatus). Space constraints in an article of this type prevent the presentation of all the exercises involved, so information is provided here for three activities addressing the core concepts.

The investigations are built around the HHMI Holiday Lecture entitled “Fossils, Genes, and Embryos,” delivered by David Kingsley (Howard Hughes Medical Institute, 2005). Although the lecture can be shown in its entirety at the beginning of the lab session, it is preferable to show video clips in small increments throughout the exercises in order to foster the inquiry-based nature of the activities.

The Fossil Record of Sticklebacks

Sticklebacks are small fish averaging about 3–5 inches in length (Figure 1). Biologists are fascinated by sticklebacks because of their wide-ranging characteristics and because of the rich deposits of stickleback fossils that have been found around the world. One such deposit is located in Nevada, where thousands upon thousands of stickleback fossils are deposited in thin layers of diatomaceous earth. The layers can be manually separated and read almost like the pages of a book. This allows paleontologists to examine the evolution of stickleback traits over time with high chronological resolution. For example, it is possible to track both the loss and reemergence of pelvic spines at the Nevada site over the course of just 30,000 years (Bell et al., 2006).

Figure 1.

Threespine stickleback (Gasterosteus aculeatus). Courtesy of New York State Department of Environmental Conservation.

Figure 1.

Threespine stickleback (Gasterosteus aculeatus). Courtesy of New York State Department of Environmental Conservation.

Lab Activity: Analysis of Stickleback Fossils

In this first exercise, students track changes in stickleback morphology by analyzing the relative abundance of different fossil phenotypes over a specific time period. This activity can be introduced by showing students Video Clip 1 (see Additional Resources at the end of this article).

Stickleback fossils at the Nevada site come in three forms with regard to the development of pelvic spines: reduced pelvis (negligible spines), intermediate pelvis (partially complete spines), and complete pelvis (fully developed spines). Representative fossils are shown in Figure 2. Students are provided a 30-page booklet depicting 30 simulated fossil layers from the site; a sample page is shown in Figure 3 (see Additional Resources for access to the complete booklet). The layers (pages of the booklet) represent 1000-year increments, layer 1 being the oldest and layer 30 the most recent. Students count the number of stickleback fossils of each phenotype present in each layer and record their observations in a table (Table 1). Note that the values in Table 1 represent percentages of each phenotype (i.e., number of fossils of specific phenotype in layer divided by total number of fossils in layer). Students are then asked to graph the relative frequency of each fossil phenotype. This can be done by hand, or (preferably) using EXCEL (see Additional Resources for access to an EXCEL graphing template). The completed graph will look like Figure 4.

Figure 2.

Stickleback fossil phenotypes (adapted from Howard Hughes Medical Institute, 2005).

Figure 2.

Stickleback fossil phenotypes (adapted from Howard Hughes Medical Institute, 2005).

Figure 3.

Sample page from Stickleback Fossil Booklet.

Figure 3.

Sample page from Stickleback Fossil Booklet.

Figure 4.

Results of fossil phenotype analysis.

Figure 4.

Results of fossil phenotype analysis.

Table 1.
Phenotypic frequencies of stickleback fossils.
LayerReduced PelvisIntermediate PelvisComplete Pelvis
1 (oldest) 100.0 0.0 0.0 
2 81.8 9.1 9.1 
3 90.9 0.0 9.1 
4 100.0 0.0 0.0 
5 100.0 0.0 0.0 
6 83.3 8.3 8.3 
7 100.0 0.0 0.0 
8 90.0 0.0 10.0 
9 22.2 22.2 55.6 
10 13.3 20.0 66.7 
11 7.7 15.4 76.9 
12 7.1 21.4 71.4 
13 21.4 14.3 64.3 
14 50.0 7.1 42.9 
15 53.8 7.7 38.5 
16 80.0 0.0 20.0 
17 90.9 0.0 9.1 
18 100.0 0.0 0.0 
19 90.0 10.0 0.0 
20 83.3 0.0 16.7 
21 90.0 0.0 10.0 
22 100.0 0.0 0.0 
23 83.3 0.0 16.7 
24 100.0 0.0 0.0 
25 90.9 9.1 0.0 
26 90.0 0.0 10.0 
27 100.0 0.0 0.0 
28 100.0 0.0 0.0 
29 90.0 0.0 10.0 
30 (newest) 83.3 8.3 8.3 
LayerReduced PelvisIntermediate PelvisComplete Pelvis
1 (oldest) 100.0 0.0 0.0 
2 81.8 9.1 9.1 
3 90.9 0.0 9.1 
4 100.0 0.0 0.0 
5 100.0 0.0 0.0 
6 83.3 8.3 8.3 
7 100.0 0.0 0.0 
8 90.0 0.0 10.0 
9 22.2 22.2 55.6 
10 13.3 20.0 66.7 
11 7.7 15.4 76.9 
12 7.1 21.4 71.4 
13 21.4 14.3 64.3 
14 50.0 7.1 42.9 
15 53.8 7.7 38.5 
16 80.0 0.0 20.0 
17 90.9 0.0 9.1 
18 100.0 0.0 0.0 
19 90.0 10.0 0.0 
20 83.3 0.0 16.7 
21 90.0 0.0 10.0 
22 100.0 0.0 0.0 
23 83.3 0.0 16.7 
24 100.0 0.0 0.0 
25 90.9 9.1 0.0 
26 90.0 0.0 10.0 
27 100.0 0.0 0.0 
28 100.0 0.0 0.0 
29 90.0 0.0 10.0 
30 (newest) 83.3 8.3 8.3 

The graph can lead to a general discussion of fossil evidence, particularly regarding the existence of transitional forms. Video Clip 2 provides a good explanation of the results. In this particular case, transitional forms are present at moderate frequencies during the loss and subsequent reemergence of pelvic spines. Very few fossil deposits are as complete and highly resolved chronologically as the Nevada site. Thus, transitional fossils, in general, are often scarce and difficult to find.

Genetic Analysis of Stickleback Spine Reduction

Although there are several extant species of both marine and freshwater sticklebacks, the ancestral form was marine and spawned in freshwater streams. As glaciers receded at the end of the last ice age (~10,000 years ago), a large number of freshwater lakes and streams formed. Oceanic sticklebacks that invaded freshwater lakes and streams became isolated. Over time these freshwater populations diverged from their oceanic ancestors and acquired new traits (Peichel & Boughman, 2003).

For example, while almost all extant marine populations possess pelvic spines, certain freshwater populations have lost pelvic spines. Pelvic spines helped protect sticklebacks from predators in the marine environment, but these predators were not as abundant in freshwater lakes. As such, there was less selective pressure to possess spines. Indeed, predators in freshwater environments include insects that attack young sticklebacks by latching onto their spines, thus selecting for spine reduction. It is also possible that lower levels of calcium ions in freshwater lakes may have selected against the production of the calcium-rich spines.

Researchers have used existing populations of sticklebacks to uncover the genes associated with pelvic spine reduction. Although they expected this relatively complex trait to be controlled by multiple genes, artificial cross-breeding experiments involving marine and freshwater populations revealed that pelvic spines are regulated by a single gene locus (Cresko et al., 2004).

Note: Students typically perform a simulation (not presented here) of cross-breeding experiments that allow them to discover the monogenic inheritance pattern of pelvic spines for themselves. The simulation is adapted from a case-study exercise entitled The Case of the Three-Spine Stickleback (Platt & Flammer, 2007). More recently, HHMI has developed a similar simulation for their Stickleback Evolution Virtual Lab that could be used to accomplish the same purpose (Howard Hughes Medical Institute, 2013).

Lab Activity: Modeling Hardy-Weinberg Equilibrium

Students are now ready to investigate pelvic spine reduction from the perspective of population genetics. The Hardy-Weinberg principle states that allele and genotype frequencies within a sexually reproducing diploid population will remain in equilibrium unless outside forces act to disturb them. Disturbing influences may include, but are not limited to, nonrandom mating, mutations, natural selection, limited population size, random genetic drift, and gene flow.

It is important for students to understand that outside the controlled settings of the lab, one or more of these “disturbing influences” are always in effect. Thus, most populations do not remain in genetic equilibrium for long. Rather, genetic equilibrium is a hypothetical state that provides a baseline for measuring actual genetic change. Hardy-Weinberg can be used to evaluate genetic change in stickleback populations.

Students first use a web-based population genetics simulator to analyze allele and genotype frequencies within a hypothetical population at equilibrium. The advantage of computer simulations is that they can generate large amounts of data in a short time. The simulator used in this exercise is the PopGen simulator created by Bob Sheehy (available at http://www.radford.edu/~rsheehy/Gen_flash/popgen).

The PopGen simulator is easy to use and allows inputs for controlling a variety of variables, including population size, initial frequencies of alleles, and fitness coefficients. The simulator refers to different alleles as “A1” and “A2”. Students can assume that A1 represents the allele for pelvic spine production, while A2 refers to the allele for pelvic spine loss. Heterozygous individuals (A1A2) display intermediate spines (Figure 2). In order to demonstrate genetic equilibrium for this single gene locus, students run an initial simulation using the following assumptions:

  • Number of populations = 1

  • Population size = 10,000

  • Number of generations = 100

  • Initial frequency of A1 (p) = 0.8

  • Genotype fitness: A1A1 = 1, A1A2 = 1, A2A2 = 1

This will result in a graph similar to Figure 5. Since the fitness of each genotype was set at 1.0, the population remains in genetic equilibrium.

Figure 5.

Allele and genotype frequencies in a population at equilibrium (PopGen simulator).

Figure 5.

Allele and genotype frequencies in a population at equilibrium (PopGen simulator).

Fossil data from the first exercise, however, demonstrate that phenotypic frequencies of sticklebacks changed over time. Phenotypic changes are often the result of natural selection operating on a population. Have students think about and discuss how natural selection might influence allele frequencies within a population of stickleback. Showing Video Clip 3 can help facilitate the discussion.

Lab Activity: Effect of Natural Selection on Stickleback Populations

Students can use the PopGen simulator to analyze how natural selection could have disturbed allele frequencies within the isolated marine stickleback populations. Recall that marine populations became isolated in freshwater streams and lakes following the last ice age. Since marine populations typically have spines, assume that the frequency of the A1 allele in the starting population is quite high (p = 0.95), while the frequency of the A2 allele is quite low (q = 0.05). Once isolated in fresh water, however, selection would have favored the reduced-pelvic-spine phenotype, corresponding to the A2A2 genotype. Therefore, assume the following fitness coefficients: A1A1 = 0.75, A1A2 = 0.85, and A2A2 = 1. Have students run the simulator for a single population of 10,000 individuals for 100 generations. Their results should be similar to Figure 6.

Figure 6.

Effects of natural selection on allele and genotype frequencies in a stickleback population (PopGen simulator).

Figure 6.

Effects of natural selection on allele and genotype frequencies in a stickleback population (PopGen simulator).

The bottom graph in Figure 6 strongly resembles the graph of fossil phenotypes in Figure 4. Discuss with students why the two graphs are similar. Although the computer simulation results are hypothetical, they demonstrate how natural selection and population genetics could account for the phenotypic changes observed in the fossil record.

Molecular Genetic Analysis of Stickleback Evolution

By this point, students have looked at pelvic spine reduction in sticklebacks from the perspectives of the fossil record and population genetics. They are now ready to turn their attention to molecular genetics and attempt to identify the precise genetic mutation responsible for this evolutionary event.

The gene that controls pelvic spines is called Pitx1 (Shapiro et al., 2004). Curiously, this gene codes for a protein that does not make up any part of the pelvic spine. Rather, the Pitx1 protein binds to the regulatory DNA of other genes to turn them either “on” or “off.” Some of these genes probably do make up parts of the pelvic spines.

Both marine and freshwater sticklebacks produce Pitx1 protein. It would seem reasonable then that Pitx1 proteins from marine sticklebacks and Pitx1 proteins from freshwater sticklebacks are somehow different. If so, there would likely be a difference between the coding regions of the Pitx1 genes of marine and freshwater fish. This prediction, however, turns out to be wrong. Scientists were able to determine the DNA base sequences of the coding regions of the Pitx1 genes from both marine and freshwater sticklebacks and found that the two sequences were identical. Perhaps the Pitx1 gene is present but not expressed (i.e., no Pitx1 protein produced) within certain tissue regions of the different phenotypes. Have students think about and discuss how this could be possible.

Genes consist of many more bases than are needed to simply code for mRNA. Many of these extra bases make up the regulatory DNA (Figure 7). In the diagram, the circle, square, and triangle represent three separate segments of regulatory DNA. These regulatory sequences act as “switches” for controlling the differential expression of the gene. A switch is “activated” when a transcription factor binds to it. This, in turn, activates transcription of the gene's coding region and the gene is expressed. The presence of transcription factors can vary in different tissues, allowing genes to be differentially expressed in different tissue types.

Figure 7.

Model of genetic regulation by “gene switches.”

Figure 7.

Model of genetic regulation by “gene switches.”

Researchers investigating Pitx1 used molecular-dye techniques to reveal the presence of Pitx1 mRNA, thus revealing where the Pitx1 gene is expressed in developing fish embryos (Shapiro et al., 2004). The researchers compared the pattern of Pitx1 mRNA expression between marine and freshwater stickleback embryos. The results are shown in Figure 8. Notice that the Pitx1 gene is expressed in the olfactory region and jaw of both marine and freshwater fish (panels C and E), suggesting that Pitx1 is important for the development of additional features besides pelvic spines. Additionally, the gene is expressed in the pelvic region of marine fish (panel D), but not in freshwater fish (panel F). This accounts for the lack of pelvic spines in the freshwater phenotype. Video Clip 4 shows how Pitx1 expression is differentially regulated by gene switches.

Figure 8.

Pitx1 mRNA expression in marine and freshwater sticklebacks (adapted from Shapiro et al., 2004).

Figure 8.

Pitx1 mRNA expression in marine and freshwater sticklebacks (adapted from Shapiro et al., 2004).

The specific mutation that causes the transcription of Pitx1 mRNA to be “turned off” in the pelvis is located in the piece of regulatory DNA that normally activates Pitx1 transcription in the pelvis (Figure 9). The regulatory switches that control transcription of the gene in the olfactory region and jaw are unaffected.

Figure 9.

Model of Pitx1 gene regulation in marine and freshwater sticklebacks.

Figure 9.

Model of Pitx1 gene regulation in marine and freshwater sticklebacks.

Lab Activity: Identifying Gene Switch Mutations Using PCR

The hypothesis that a mutation exists in the pelvis regulatory region of the Pitx1 gene in freshwater fish was confirmed by mapping the chromosomal regions associated with Pitx1 expression in both marine and freshwater fish, and then amplifying these regions by means of polymerase chain reaction (PCR). The amplified fragments from marine (complete pelvis) and freshwater (reduced pelvis) fish could then be compared for differences in size, potentially stemming from deletion or insertion mutations in the regulatory regions.

Researchers studied several populations of pelvic-reduced freshwater sticklebacks throughout the northern United States and Alaska. DNA genotyping of the populations identified nine different deletions overlapping in a 488-bp region, resulting in pelvic reduction (Chan et al., 2010). Results of the research are shown in Figure 10.

Figure 10.

Genotyping of freshwater pelvic-reduced freshwater sticklebacks from North American lakes (adapted from Chan et al., 2010).

Figure 10.

Genotyping of freshwater pelvic-reduced freshwater sticklebacks from North American lakes (adapted from Chan et al., 2010).

In this final lab exercise, students amplify Pitx1 regulatory regions corresponding to jaw and pelvis development in marine, freshwater, and F1 hybrid sticklebacks to discover the genetic mutation outlined above for themselves. The exercise is performed using components from Edvotek kit no. 371 – PCR-based DNA Fingerprinting (http://www.edvotek.com/). (Note: Edvotek kit 371 is designed to teach DNA fingerprinting by utilizing a crime scene investigation. When using this kit to generate results for this lab, all three jaw samples correspond to suspect 3, marine pelvis = suspect 2, freshwater pelvis = suspect 1, and hybrid pelvis = suspect 4. If PCR equipment is not available, students can simply be provided a picture of the results. Alternatively, if only electrophoresis equipment is available, students can separate preamplified DNA fragments available by contacting Edvotek directly.)

Once the samples are amplified and separated by gel electrophoresis, students obtain a gel similar to the one in Figure 11. Have students think about and discuss the pattern of amplified fragments revealed by the gel. At this point they should be able to recognize that a deletion of some sort in the pelvic regulatory region of Pitx1 (evidenced by the smaller fragments) accounts for the reduction of pelvic spines in freshwater sticklebacks.

Figure 11.

Electrophoretic separation of simulated PCR-amplified Pitx1 regulatory DNA.

Figure 11.

Electrophoretic separation of simulated PCR-amplified Pitx1 regulatory DNA.

Discussion

The primary advantage of these exercises is that students investigate a real evolutionary event using a variety of techniques and data. This serves to unify the sometimes disparate lines of evolutionary evidence. The complete set of exercises usually requires two lab periods. Ideally, students work with a partner, although cost and equipment availability may require the PCR and electrophoresis steps to be done in larger groups or as an entire class. Students will also need access to computers for performing the PopGen simulation and EXCEL graphing steps. The fossil layers can be printed in booklet form for kinesthetic effect, but they can also be read as a PDF from a computer or tablet. Freshwater sticklebacks are readily available in certain parts of the country where they are often used as minnow bait. Setting up a tank of sticklebacks in the lab could increase student interest and make the exercises all the more realistic.

Although students can view the entire HHMI lecture regarding stickleback evolution at the beginning of the exercises, presenting the activities as a logical series of structured inquiries allows for increased student engagement and more opportunity for critical thinking. Assessment can consist of quiz and exam questions drawn from the lab content, as well as critical essay questions regarding experimental design and data interpretation.

The reduction or complete loss of the pelvis and hind limbs has occurred not just in sticklebacks, but repeatedly in the course of vertebrate evolution. It is seen in snakes, legless lizards, whales, dolphins, and manatees. Based on the previous exercises involving sticklebacks, students should be able to make a reasonable hypothesis regarding the evolutionary and genetic basis of pelvic reduction in other species.

Explanations for events such as pelvic spine loss have led to novel insights into the evolutionary process. The discovery of HOX genes coupled with the sequencing of complete genomes has led to a paradigm shift in both genetics and evolutionary biology. Mutations in regulatory DNA that affect embryological development are of particular interest to biologists because these mutations can have a dramatic influence on phenotypic form and function. Evolutionary-developmental biology (evo-devo) compares the developmental processes of different organisms to determine the ancestral relationship between them and to discover how developmental processes evolved. Evo-devo demonstrates that evolution alters developmental processes to create new and novel structures from existing gene networks. It is a fascinating new field, and one that students will enjoy exploring further.

Additional Resources

References

References
Bell, M.A., Travis, M.P. & Blouw, D.M. (
2006
).
Inferring natural selection in a fossil threespine stickleback
.
Paleobiology
,
32
,
562
577
.
Chan, Y.F., Marks, M.E., Jones, F.C., Villarreal, G., Jr., Shapiro, M.D., Brady, S.D. et al. (
2010
).
Adaptive evolution of pelvic reduction in sticklebacks by recurrent deletion of a Pitx1 enhancer
.
Science Magazine
,
327
,
302
305
.
Cresko, W.A., Amores, A., Wilson, C., Murphy, J., Currey, M., Phillips, P. et al. (
2004
).
Parallel genetic basis for repeated evolution of armor loss in Alaskan threespine stickleback populations
.
Proceedings of the National Academy of Sciences USA
,
101
,
6050
6055
.
Howard Hughes Medical Institute
(
2005
).
Fossils, genes, and embryos
. Available online at http://www.hhmi.org/biointeractive/fossils-genes-and-embryos.
Howard Hughes Medical Institute
(
2013
).
Stickleback evolution virtual lab
. Available online at http://www.hhmi.org/biointeractive/stickleback-evolution-virtual-lab.
Peichel, C.L. & Boughman, J.W. (
2003
).
Quick guide to sticklebacks
.
Current Biology
,
13
,
R942
R943
.
Platt, J.E. & Flammer, L. (
2007
).
Case of the threespine stickleback
. Available online at http://www.indiana.edu/~ensiweb/lessons/stickleback.html.
Shapiro, M.D., Marks, M.E., Peichel, C.L, Blackman, B.K., Nereng, K.S., Jónsson, B. et al. (
2004
).
Genetic and developmental basis of evolutionary pelvic reduction in threespine sticklebacks
.
Nature
,
428
,
717
723
.
Sheehy, R. (
2013
).
Population genetics simulator
. Available online at http://www.radford.edu/~rsheehy/Gen_flash/popgen/.