True breeding strains. The Drosophila mutants shibire (stock #2248) and Lobe (stock #324) were obtained from the Bloomington Drosophila Stock Center at Indiana University, Bloomington. Additionally, white (87 W 6553), eyeless (87 W 6592), apricot (87 W 6562), wild type (87 W 6621), and Bar (87 W 6551) flies were obtained from Ward’s Biology and Chemistry (Rochester, NY). The eosin mutants (white-eosin 17-2240, no longer available; obtain eosin mutants from the Bloomington Stock Center) were obtained from Carolina Biological Supply (Burlington, NC). All flies were cultured in Ward’s instant Drosophila medium or Carolina formula 4-24, kept at 25^oC, and transferred to new vials at 1- to 2-week intervals to be kept as true breeding strains.  

Drosophila crosses. For the lab activity, heterozygous apricot-white and eosin-white female Drosophila were also needed. Four weeks before the activity, a vial of true breeding eosin flies was mixed with a vial of true breeding white flies and allowed to mate indiscriminately. Every week, all adults that were in the vials were transferred to a new food vial. After 4 weeks, the flies that were present in the vials represented a mixture of true breeding white, true breeding eosin, and heterozygous eosin females. Four weeks before the activity, a vial of true breeding apricot flies was mixed with a vial of true breeding white flies and allowed to mate indiscriminately. Every week, all the adults that were present in the vial were transferred to a new food vial. After 4 weeks, the flies that were present in the vials represented a mixture of true breeding white, true breeding apricot, and heterozygous apricot female flies.   

Fly manipulation. Flies were manipulated as described in the Carolina Drosophila manual (4).  Approximately 30 adult wild type flies and the shibire Drosophila true breeding adult flies were transferred separately to clean vials free of media.  These flies were used for the conditional demonstration.  

For all other Drosophila analysis, flies of the appropriate true breeding strain or cross were transferred to labeled clean vials, free of food and netting. Flies were distributed to students, and the students anesthetized them using Carolina Flynap and Flynap wands until the flies were no longer moving. These flies were used for the remainder of the examinations.  

All anesthetized flies were analyzed under a dissecting microscope on white cards, using soft paint brushes to move the flies for sexing and phenotype examination.  

Analysis of student learning. Exemption from human subjects approval was granted by the Institutional Review Board as pursuant to Federal regulation 45 CFR Part 46.101(b)(1). Willing students filled out anonymous pre- and postlab questionnaires (3, 6). Four total lab groups were analyzed (two in fall 2006 and two in spring 2007) and the data were combined such that the questionnaires could not be assigned to a lab section. The mean for each question and the standard deviation were calculated using Excel. Identical pre- and postlab questions on the student self-assessment of learning were compared using Excel and a one-tailed homoscedastic student’s t test to test for significance (Table 1).  Additional survey questions (Table 2) and the opportunity to provide open-ended comments (Table 3) were included on the postlab survey. Pre- and postlab testing was also employed in the spring of 2007 to determine the extent of actual student learning; a quiz (Pre- and Postlab Quiz) was given directly before the exercise and then the same quiz was given the following week in lab. The data from the pre- and postlab quiz were analyzed for the percentage of correct answers using all of the student’s results (Table 4).  

Statement	Prelab student self-assessment mean ± SD (n = 71)	Postlab student self-assessment mean ± SD (n = 67)	P value
I understand well the concept of simple Mendelian inheritance.	4.63 ± 0.57	4.69 ± 0.50	0.23
I understand well the concept of multiple alleles.	4.00 ± 0.92	4.29 ± 0.81	0.024b
I understand well the concept of gene dosage effect.	4.14 ± 0.86	4.50 ± 0.68	0.0034b
I understand well the concept of expressivity.	4.18 ± 0.66	4.44 ± 0.65	0.010b
I understand well the concept of incomplete penetrance.	3.79 ± 0.87	4.07 ± 0.85	0.028b
I understand well the concept of protein level affecting phenotype and enzyme activity.	3.81 ± 1.06	4.22 ± 0.79	0.038b
I understand well the concept of temperature sensitivity.	4.24 ± 0.85	4.47 ± 0.82	0.049b
I feel comfortable manipulating and sexing flies.	3.47 ± 1.01	4.15 ± 1.05	8.25 x 10-5b
I feel comfortable scoring phenotypes in Drosophila.	3.33 ± 0.93	3.91 ± 1.06	4.02 x 10-4b

 

Statement	Postlab student response mean ± SD  (n = 67)
I enjoyed this lab.	3.82 ± 1.08
I would recommend that this lab exercise be kept in the genetics curriculum.	4.01 ± 1.06
This lab exercise made me think.	4.08 ± 0.97
This lab exercise fit in well with the curriculum of the lecture.	4.26 ± 0.92
I hate this particular lab.	2.39 ± 1.21
This lab was an active process for me.	4.12 ± 0.94
I learned something from this lab.	4.33 ± 0.73
This lab made me ask questions, such as “why does this fly look this way?”	4.26 ± 0.86

Responses that indicate the learning objectives were achieved

“This lab is really interesting and fun!  Definitely keep it for years to come!”

“Good lab.”

“This was great practice for sexing and manipulating flies.”

“This lab was very helpful in distinguishing the different phenotypes.  Did not enjoy last weeks [sic] lab [setting up Drosophila parental crosses, vestigial ebony x wild type and sepia ebony x wild type, to obtain and analyze F1 and F2 generation].  It was kind of confusing but now that we know what it was that we were looking at it makes much more sense.”

“I enjoy lab’s [sic] like this that are not “cookbook” lab’s [sic], but isn’t something better than flie’s [sic] we could study?”

“Differences in shades of color was [sic] much more frustrating than I thought.”

Responses that indicate the learning objectives were not achieved

“Lobe eye was confusing.”

“Counting the flies was confusing.”

“Its [sic] a lab where we didn’t do anything but observe; thoes [sic] are not my favorite labs.”

“I believe specific instructions on what we needed to do will be better.  We were not sure which ones to count, and why it was necessary to count.”

 

Question number and topic	Prelab quiz percentage of correct answers	Postlab quiz percentage of correct answers	Change in percentage of correct answers from pre- to postlab	P value
1. conditional allele	83.3	97.2	13.9	0.0736
2. gene dosage effect	80.5	88.9	8.4	0.6171
3. gene dosage effect	52.8	94.4	41.6	0.0012 b
4. relate gene dosage effect to protein levels	50	52.8	2.8	0.7728
5. conditional allele	86.1	91.7	5.6	0.4497
6. multiple alleles	27.8	77.8	50	<0.0001 b
7. multiple alleles	72.2	83.3	11.1	0.4497
8. expressivity	72.2	100	27.8	0.0077 b
9. incomplete penetrance	69.4	86.1	16.7	0.0704
10. dosage compensation	8.3	44.4	36.1	0.0019 b
11. dominant gain of function mutation	38.9	66.7	27.8	0.0055 b
12. gene dosage effect	36.1	86.1	50	0.0002 b
13. dosage compensation	2.8	47.2	44.4	0.0002 b
14. multiple alleles	41.7	80.6	38.9	0.0055 b
Mean	51.6	78.4	26.8

^a Students (n = 36) were given the same quiz on the laboratory concepts directly before (pre-) and 1 week after (post-) the lab exercise to assess the extent of student learning from the exercise.   Statistical significance was determined using a McNemar’s test for paired enumerated data, two tails. 

^b P < 0.05.

The data in Table 4 were further analyzed for a statistically significant difference between a pre- and postlab exercise quiz using a McNemar’s test for paired enumeration data using Graphpad software (http://www.graphpad.com/quickcalcs/McNemar1.cfm) (Table 4), removing from the analysis those students who did not take both quizzes (one did not take the posttest and a different student did not take the pretest). The McNemar’s test is used to determine if there is a statistically significant difference in paired dichotomous data. This statistical test is best used for data in which the results are in a yes-no format. Each student’s quiz results are described as either correct on both pretest and posttest, correct on only pretest, correct on only posttest, or correct on neither pre- nor posttest. The null hypothesis is thus that those who did not know the answer before the exercise will not know the answer on the posttest.  

Student prelab preparation. Prior to lab, students were given a reading assignment of the relevant chapters in the textbook and the laboratory procedures (Lab Handout). Then students were required to take an online prelab quiz through the BlackBoard course shell to ensure that they had read and understood the assignment. Students were also given a prelab self-assessment of their knowledge and abilities (Table 1) and a prelab quiz (spring 2007 only).  This assessment was given during the lab time but before the activity commenced. 

In lab exercises.  (i)  Conditional allele demonstration.  Conditional alleles are those that have a wild type phenotype in one environmental condition but show a mutant or variant phenotype in another environmental condition (1). Students analyzed an obvious and reversible temperature sensitive mutant of Drosophila, speculating about the protein function, activity, and structure at permissive and nonpermissive temperatures. shibire is a mutation of dynamin in which, at the restrictive temperature, the flies are paralyzed because they cannot complete synaptic transmission (10). This mutation is fully reversible, and the flies regain movement once they are returned to the permissive temperature. The conditional allele demonstration of the shibire mutation is a variation of a demonstration by A. Bejsovec (http://flystocks.bio.indiana.edu/Browse/misc-browse/Bejsovec.htm). True breeding homozygous shibire and wild type flies were transferred to clean vials and stored at room temperature. Students were allowed to watch the fly activity in the vials to determine that both strains were alive and capable of movement. The instructor then grasped each vial tightly in her hand as the students used a watch with a second hand to determine how long it took for the shibire mutants to cease movement and fall to the bottom of the vial. Once the shibire mutants were no longer moving, both vials were placed on the counter at room temperature and the students measured how long it took for the shibire mutant flies to regain movement. After the demonstration, students were asked questions about the function of the dynamin protein and about the structure and activity of the dynamin protein in the shibire mutant at the restrictive and nonrestrictive temperatures in order to stimulate their thinking about the molecular process and protein function of temperature sensitive mutants.  

(ii)  Phenotype demonstrations for other extensions of Mendelian inheritance. For this exercise, separate dissecting microscope stations were set up for each true breeding strain or cross. Each lab group was given one or two labeled vials with at least 40 flies in each vial, free of food and netting, to anesthetize. Students used Flynap to anesthetize the flies,  placed them under a dissecting microscope, and clearly labeled the station as to the true breeding strain or cross. Each student looked at all of the different true breeding strains and crosses in “round robin” style. Students compared all mutant flies to the wild type Drosophila strain. 

(a)  Variable expressivity and incomplete penetrance. Phenotypes in mutants can vary from individual to individual such that all members of a group will have the same genotype but will vary in their phenotype. Variable expressivity is the phenomenon in which all individuals of a given genotype will show differing degrees of a phenotype. Some dominant phenotypes can even skip generations, in an inheritance pattern known as incomplete penetrance (1). To illustrate variable expressivity and incomplete penetrance, students analyzed eye size phenotypes. Students examined true breeding Drosophila mutants of Lobe and eyeless and compared the eye phenotype of shape, size, and facet number to a wild type Drosophila strain. The recessive mutation, eyeless, shows variable expressivity. Homozygous eyeless flies have eyes that range from almost wild type in appearance to almost completely absent. The most striking observation was that eyeless flies can have one eye that has an almost wild type phenotype, while the other eye is almost completely absent; thus showing variable expressivity even within the same organism. The Lobe phenotype is more subtle and also demonstrates incomplete penetrance. The most common Lobe phenotype is a loss of the ventral side of the eye (2); however, the phenotype can be more or less severe as the allele also demonstrates variable expressivity. Upon first inspection of the Lobe mutants, most students indicated that all flies examined had a wild type phenotype. Only after explanation by the instructor and reexamination, did the students see the eye size phenotype of Lobe mutants. Students were asked questions about the protein function in the mutant fly strains. In order to reinforce the idea that traits are governed by protein function, students were asked in postlab exercise questions to describe how they thought protein function could be affecting the phenotype in terms of variable expressivity and incomplete penetrance. 

(b)  Multiple alleles. Many genes have more than the binary two alleles; this inheritance is termed multiple alleles (1). Students studied eye color mutants in order to understand multiple alleles. The X-linked white gene exhibits alleles of eosin, apricot, and white. Students examined eye color in these three mutants as compared to wild type and were asked in postlab questions to deduce a possible explanation of protein function for each allele.  

(c)  Dosage compensation. Dosage compensation is the phenomenon in which the gene expression of most X-linked genes does not vary between hemizygous males and homozygous females (1). In Drosophila, the expression of most X-linked genes is doubled in the male fly to equal that in the homozygous female fly. Students examined wild type eye color and the apricot eye color allele to demonstrate that both the hemizygous male and the homozygous female have the same eye color. In postlab questions, students were asked about the phenomenon of dosage compensation and how it occurs at the molecular and protein expression levels.  

(d)  Gene dosage effect. Gene dosage effect is an inheritance pattern in which the number of copies of an allele affects the intensity or severity of the phenotype (1). For understanding of gene dosage effect, students examined eye mutants, including two eye color mutants and one eye shape mutant. For analysis of all gene dosage effect mutants, students were asked a series of postlab questions about the phenotype of each mutant and the relationship of the protein function and levels to the phenotype. In postlab questions, students were encouraged to explore how protein levels in gene dosage effect mutants would lead to different phenotypic consequences.  

The mutant alleles eosin and apricot are both X-linked genes that demonstrate gene dosage effect. For eosin eye color in Drosophila, the true breeding wild type females have twice as much pigment and a darker eye color than the hemizygous males do. Heterozygous females (eosin-white heterozygotes) have an eye color as intense as the males but  less intense than the homozygous females (1). The phenotype of apricot eye color in Drosophila is more complicated. The apricot allele demonstrates dosage compensation as well as gene dosage effect. In males with the apricot allele, the eyes are peach in color. In the homozygous females, the eye color is the same color as the males. The heterozygous female (apricot-white heterozygote) has a lighter eye color than the homozygous female or the male. Students analyzed true breeding white, eosin, and apricot Drosophila to inspect eye color. Students also were required to sex these flies.  Once students were familiar with the eye colors of the true breeding flies, they examined the eye color phenotype in the Drosophila crosses, which were a mixture of males and homozygous and heterozygous females. Students noted eye color and sex of flies in each cross for a minimum of 25 flies, looking for the heterozygous females in each case. For both the eosin and apricot true breeding strains, students were asked to describe in postlab questions the eye color in males and females in terms of the relationship to protein level and function. Students also had to describe why in true breeding apricot flies the males and females had similar eye colors, but in eosin the males had lighter eyes than the females.    

The Bar allele is another example of an allele that is governed by gene dosage effect. This allele is also an example of an X-linked dominant gain-of-function allele. In true breeding bar eye flies, the facet number is reduced as compared to wild type, with the females having fewer facets than the males. Students analyzed a true breeding stock of bar eye flies to study the gene dosage effect of this allele. Students described the protein levels as they related to the differing eye phenotypes in males and females. In postlab questions, students also were asked about how they thought the Bar allele was working as a gain-of-function mutation, in order to help students link the allele with protein function.  Students were also questioned about the formation of wild type revertants, males with typical female Bar phenotype and Ultrabar females (1) and what the protein function and levels would be in these cases.  

RESULTS AND DISCUSSION  

Students used mutants of Drosophila to examine and contemplate extensions of Mendelian inheritance. They were asked a series of postexercise questions about each mutant, focusing on the level, function, and/or structure of each protein in each mutant to direct their thinking towards the relationships between the protein coded for by each mutant allele and the phenotypes seen.   

Assessment. Assessment was determined via pre- and postlab questionnaires on students’ skill and knowledge (Table 1) and by a postexercise questionnaire on the students’ perception of the activity (Table 2) (3, 6). Pre- and postlab  answers were analyzed using a one-tailed homoscedastic student’s t test (Table 1). Students were asked a series of survey questions pre- and postlab. Questions addressing students’ self-assessment of knowledge and skills indicated that the students thought that they understood the concepts covered by the lab better after the lab exercise and felt that their skills of sexing, manipulating, and scoring phenotypes in Drosophila improved (Table 1). In all cases, except for the question concerning simple Mendelian inheritance, there was a significant difference between the pre- and postlab self-assessment of knowledge and skills, indicating that the students had in fact felt as though they had learned from the exercise. This lab was a worthwhile exercise to increase students’ self-assessment of their knowledge and their actual understanding of the topics.  

This lab was also successful in building students’ fly manipulation skills. This exercise falls in our lab curriculum after an initial Drosophila inheritance lab in which the students set up crosses to generate the F₁ generation. This lab further enhanced the students Drosophila manipulation skills. According to the students’ survey answers, they did feel more comfortable manipulating and sexing flies after performing the exercise (prelab 3.47 ± 1.01 compared to postlab 4.15 ± 1.05, P value of 8.25 x 10^-5). This lab also helped the students feel more comfortable in scoring Drosophila phenotypes (prelab 3.33 ± 0.93 compared to postlab 3.91 ± 1.06, P value of 4.02 x 10^-4). The responses to the pre- and postlab survey indicate that the learning objectives of increased ability in fly manipulation are being met.  

Additional postlab survey assessment indicated that the students felt the lab was a worthwhile exercise and should be kept in the curriculum. The responses to the postexercise survey indicated that students felt they had learned something, the lab exercise fit in well with the curriculum, the exercise was an active process, and the exercise made them ask questions and think; all responses to these survey questions were at least a 4 (agree slightly) on a 5-point Likert-type scale (Table 2). As a part of the postexercise survey, students were able to write additional comments about the lab, which are detailed in Table 3. These comments fall into two main categories, those that indicate the learning goals were achieved and those that indicate that the lab did not achieve the learning goals.  The negative comments included that the lab or a particular phenotype was confusing. This lab can be confusing because the phenotypes of Lobe and eyeless are variable and can be subtle. One of the learning objectives is to acquaint the students with careful observation of the phenotypes. The other main complaint was that the counting was confusing or that they did not understand why they needed to count the flies. I specifically included the instructions about counting in the lab manual to ensure that the students looked at more than one fly per phenotype, as many of the phenotypes vary from fly to fly. Most student comments indicated that the lab was successful in its learning objective of obtaining skills of sexing and manipulating flies and scoring nonbinary phenotypes.  

Additional student assessment was employed to determine if in fact the students were learning from the exercise. Pre- and posttesting was employed to determine if the exercise taught the students about extensions of Mendelian inheritance (Table 4). Students were asked a variety of questions about the topics covered in the lab exercise in a quiz given directly before the exercise and 1 week after the exercise (Pre- and Postlab Quiz). In all cases there were more correct answers to the questions after the exercise than before the exercise. Statistically significant increases, as analyzed by McNemar’s test of paired enumeration data, were seen in students’ understanding of gene dosage effect, multiple alleles, variable expressivity, dosage compensation, and dominant gain-of-function mutations. For questions 1, 2, and 5 there was not a statistically significant increase from the pre- to the posttest, perhaps because in each case over 86% of the students had the correct answer the first time. For question 4, students were to relate protein levels to gene dosage, and although there was an increase in the percentage of correct answers, it was not statistically significant. For question 9, the increase from pre- to postexercise was almost statistically significant (P = 0.0704), perhaps because the question asked about incomplete penetrance and in the answers to the question several students confused this topic with incomplete dominance. Anecdotal experience indicates that students in our genetics course lack precision in their language, confusing topics like sex limited with sex influenced inheritance and transcription with translation. Question 7 asked about the phenotypic result of a cross between an eosin fly and a wild type fly. All of the students (six total), who answered incorrectly on this question in the posttest, answered that the resulting offspring would be heterozygous eosin, indicating that they did not remember wild type was dominant over eosin. More prelab lecture and postlab discussion should resolve the remaining issues with this lab. Pre- and postexercise testing indicated that for eight of fourteen quiz questions there was a significant increase in students’ understanding, and in all cases of individual quiz questions, the percentage of correct answers on the quiz increased from pre- to postexercise (Table 4).  

CONCLUSION  

This lab exercise is a novel discovery-based method to demonstrate one concept in genetics: extensions of Mendelian inheritance. The objective was to allow the students to examine phenotypes in Drosophila that exhibited an extension of Mendelian inheritance and could be difficult to discern. This lab was developed to force the students to examine traits in terms of protein function and to give them more experience in manipulating, sexing, and scoring phenotypes in flies. Many of the phenotypes were variable and subtle; thus, this lab also forces the students to deliberately and slowly analyze the phenotypes, as opposed to moving quickly through the exercise, not truly paying attention to the concepts or recording the data to the fullest extent possible.  The purpose of this exercise was to force students to spend some time analyzing flies so they could become comfortable with fly phenotype analysis and manipulation. Students should take responsibility for their education and knowledge acquisition; through discovery-based hands-on activities, the instructor can direct them to see the inherent value in their own self-directed learning.  

This lab exercise will be easily adaptable to many genetics teaching labs, as many instructors of genetics are already familiar with Drosophila husbandry and manipulation. The stocks are easily accessible and grow robustly in the lab setting with a minimum of care. The topics covered in this lab are covered in most traditional genetics lectures but not usually in labs. This lab exercise will be an active hands-on learning activity for the genetics laboratory. This exercise will give students practice and build confidence in their abilities to manipulate and sex flies. Examining Drosophila mutants with these phenotypes and answering questions about the phenotypes in terms of protein levels and function should lead to enhanced student understanding of these extensions of Mendelism. Working with phenotypes that are not binary in nature will help strengthen students’ laboratory analysis skills, build self-confidence, and should move students forward in their cognitive development, such that they are confident that they can ask and answer questions that may not have “black and white” answers. 

ACKNOWLEDGMENTS 

Special thanks go to Charles E. Deutch for initial discussion about this activity and to Thomas M. Cahill for initial discussions about statistical analysis using Excel.   I also wish to acknowledge the assistance of Roger L. Berger for directions on statistical analysis and the use of the McNemar statistical test.  Thanks also to Kelli L. Bell, the instructor of the second lab section, for help with the survey and quiz distribution and collection and for discussions on how to run the lab more effectively.  Special thanks go to the students in the LSC348 Fundamentals of Genetics lab in fall 2006 and spring 2007 for participating in the assessment of this exercise. REFERENCES