NATURAL VARIATION AND GENE EXPRESSION IN THE MALE COURTSHIP PROGRESSION OF DROSOPHILA MELANOGASTER
BY
ELIZABETH ANNE RUEDI
B.A., Lake Forest College, 2001
DISSERTATION
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biology in the Graduate College of the University of Illinois at Urbana-Champaign, 2007
Urbana, Illinois
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3290367
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
® UMI
UMI Microform 3290367 Copyright 2008 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C e r t i f i c a t e o f C o m m i t t e e A p p r o v a l
University of Illinois at Urbana-Champaign Graduate College
July 2, 2007
We hereby recommend that the thesis by:
ELIZABETH ANNE RUEDI
Entitled:
NATURAL VARIATION AND GENE EXPRESSION IN THE MALE COURTSHIP PROGRESSION OF DROSOPHILA MELANOGASTER
Be accepted in partial fulfillment of the requirements for the degree of:
Doctor of Philosophy
Signatures:
— ____ Director of Research - Kimberly A. Hughes Head o/Qepart/ient Jeffrey - D. Brawn
Committee on Final Examit/afti
iberly A. Hughes Commitu ’.mber - Hugh M. Robertson
- t r -
Co} M em ber - Gene E. Robinson Committee M ember - Andrew V. Suan/z
Committee Member - Committee Member -
* Required for doctoral degree but not for master’s degree
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. © 2007 by Elizabeth Anne Ruedi. All rights reserved.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Abstract
Natural genetic variation in the temporal sequence of male courtship behavior can
be one target of selection. The goals of my dissertation research were to identify
variation in the sequence of male behavior using a natural population, and investigate the
genetic underpinnings of this variation. I developed a new, high-throughput method of
male courtship analysis; this method focuses on the progression of male courtship over
time (MCP), and is more sensitive than previously used metrics of measurement. I used
this method to examine differences in MCP between males from full-sibling families,
using either virgin or previously mated females as targets. There were significant genetic
differences in MCP for families placed with either female target, and hence there was
natural genetic variation for MCP in that population.
To investigate the genetic underpinnings of this behavioral variation, I examined
previously identified male courtship candidate genes for alterations in expression due to
changes in age or experience, and then investigated if the previously identified candidates
were also involved in natural behavioral variation. Candidate courtship genes identified
from previous research were examined for non-genotypic changes in gene expression
using genetically identical males of different ages, with different levels of exposure to
females. I found that several of these genes had altered expression due to age
(reproductive maturity) but no genes had alterations due to experience with a female.
I then performed a microarray analysis on reproductively mature males from
families with the most extreme differences in MCP from the first experiment. This
allowed me to determine which genes exhibited differential baseline expression
associated with variation in MCP. I identified fourteen candidate genes with differential
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. expression, and none of these had been previously associated with male courtship
behavior. I suggest potential roles for these fourteen new candidates, and a factor pattern
analysis revealed that several of them might be co-regulated. The culmination of this
research brings us one step closer to understanding the genetic basis for natural
behavioral variation in male courtship; this information can now be used to determine
what sequence differences are associated with expression differences in candidate genes,
and how selection might be acting on this variation.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements
This research is the culmination of six years of hard work. My time here at the
University of Illinois has been the time of my life. I have enjoyed the challenges and the
friendship, and my time in the lab. There are many people who have worked with me
and supported me; I will try to mention them all, but if I leave someone out it is entirely
unintentional. I would like to thank my advisor, Kim Hughes, for challenging me and
assisting me. Thanks to my thesis committee, Hugh Robertson, Gene Robinson, and
Andy Suarez, for their time and helpful comments. Thank you to Anne Houde for giving
me the encouragement and advising that it took to get me here in the first place. I would
like to thank Alison Bell for her stimulating conversations and suggestions for my first
manuscript. Thanks to the undergraduates who have helped me over the years, including
Karin Stokke, Rachael O’Callahan, Gavin Davis, and Amy Peterson. I would have never
survived without Melissa Reedy, the best laboratory assistant, fly wrangler, and friend.
Thanks to Silvia Remolina, Katelyn Michelini, Ashley Johnson, Rose Reynolds, and
Jenny Dmevich for being lab ladies and having fabulous ladies nights. Jenny has also
been my mentor in countless ways; thanks to her for helping me through the microarray
protocols. Thanks to Tom Newman for his assistance with qRT-PCR, extracting high-
quality DNA, and other molecular processes. Thanks to Rodrigo Valarde for helping me
troubleshoot.
I can never begin to say enough about Rose Reynolds, Amy Toth, Lynn
Anderson, and Fernando Miguez. Not only are they some of my best friends, they have
helped me survive graduate school. Whenever I needed help or comments or support or a
joke, they have been there. I hope it continues to be the same for the rest of my life.
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Thanks to my other ladies, all of whom have made huge impacts on my life at
some point during this experience: Jan Murray (winner of the best cook, support system,
and Canadian awards), Andrea Gschwend, Rachel Knepp, Lisa Ainsworth, Amy Cash,
Lydia Wraight, Corey Tarwater (winner of the best dance partner in the world award),
Molly Tranel, Eli Saetnan, Moni Berg-Binder (aka Binderberg), Jinelle Hutchins, Orla
Dermody, Maggie Prater, Melissa Chipman, Karen Cavey Rowe, Kelly Anderson, Maria
Villamil, and Yulia Tkachuk.
Thanks to my guys, who have kept me sane, made me laugh, hugged me, and let
me drink whiskey: Ian Kirwan, Paul Nabity, Paul Henne, Richard Webster, Brad
Cosentino (winner of the best movie watching partner award), Nano Tellez, Andy
Leakey, Dylan Maddox, Jeff Foster, Ken Reynolds (winner of the best baker award),
Matt Rambert, Sameer Bharwani, and Timmy Daugherty (winner of the best little brother
in the world award).
Most importantly, thank you to my family. Thank you to Melody Glaub for
giving me life, thank you to Marilyn and Mel Franklin for always being there, thanks to
Gagy and Grandpa for supporting me and making fly jokes, thanks to Grandpa and
Grandma Ruedi (I miss you). But especially, thank you to my parents. Mom and Dad,
you have been there through everything, supporting me through it all, giving me love,
encouragement, and confidence in my capabilities (even when I had lost it). I love you.
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table of Contents
Chapter 1: Overview of male courtship in the fruit fly, Drosophila melanogasler...... l
Chapter 2: Natural variation in the courtship progression of male Drosophila melanogasler...... 27 TABLES...... 50 FIGURES...... 55
Chapter 3: Alteration in expression of courtship genes due to age and experience in maleDrosophila melanogasler...... 57 TABLES...... 79 FIGURES...... 88
Chapter 4: Contribution of differential gene expression to natural variation in maleDrosophila melanogasler courtship...... 94 TABLES...... 112 FIGURES...... 116
ChapterS: Conclusion...... 118
Appendix A: Statistical tests for univariate models of single behaviors...... 121
Appendix B: Mean MCP scores for all VT families...... 122
Appendix C: Mean MCP scores for all MT families...... 125
Appendix D: Least square means of gene expression due to age, experience, or age*experience...... 128
Curriculum Vitae...... 137
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 1: Overview of male courtship in the fruit fly,
Drosophila melanogaster
Courtship behavior is vital to male fitness in many animal species because it is
necessary for successful reproduction. In species with no paternal care, mating success is
the primary determinant of adult male fitness; it can signal species identity and can be the
object of intersexual selection via direct benefits, indirect benefits, or sensory bias
modalities (Andersson 1994). Consequently, courtship success is a major fitness
component and should be under strong selection. However, like other fitness
components, this trait exhibits substantial heritable variation within populations of many
species (Stirling et al. 2002). Because variation in courtship behavior can lead to
reproductive isolation and speciation, evolutionary effects of such variation are
particularly interesting (Etges 2002; Sawamura and Tomaru 2002; Kulathinal and Singh
2004). To successfully understand the causes and consequences of this variation, we
must 1) identify the presence of genetic differences in a natural population, 2) investigate
the molecular mechanisms underlying variation in courtship behavior, and 3) identify the
evolutionary processes responsible for maintenance of variation.
My dissertation explores the first two of these questions using variation in male
courtship behavior in the fruit fly,Drosophila melanogaster. This is a well characterized
behavioral pattern in a model system, and thus is ideal for examining the behavior
genetics of natural variation. D. melanogaster male courtship behaviors are sequentially
ordered and include orienting toward the female, following/chasing the female,
performing a “courtship song” using wing vibrations, licking the female’s genitalia,
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. attempting copulation, and successful copulation (Bastock and Manning 1955). Visual,
auditory, and chemosensory cues are an integral part of successful courtship, and each
behavior plays an essential role in species recognition or assessment of the sex and
receptivity of potential partners (reviewed in Greenspan and Ferveur 2000).
Additionally, this behavioral sequence is under substantial genetic control, as revealed by
mutant screening (Hall 1994a). However, it is unknown if the genes identified by mutant
screening (or other genes) are involved in variation in male courtship behavior in natural
populations. Studies of variation in gene expression using non-mutant stocks of
Drosophila for traits such as response to alcohol (Morozova et al. 2006), response to heat
stress (Sorensen et al. 2005), and learning/memory (Dubnau et al. 2003) have produced
candidate gene lists that have very little overlap with gene lists previously identified
using mutant screening. This is true even for a study that identified candidates using
gene expression vs. mutant screening from the same source population (Dubnau et al.
2003).
Below, I review several studies investigating natural inter- and intra-specific
variation in components of male courtship behavior. I then detail the visual, auditory,
and chemosensory components contributing to male courtship behavior, as well as sex-
determination and learning. I also focus on several genes involved in these processes that
have been discovered using mutant screening; these candidate genes are important for the
production of normal courtship behavior. This is not an exhaustive list of candidates, but
rather identifies potential contributors to natural variation in behavior.
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Natural variation in Drosophila male courtship behavior
The complex courtship behavior in male Drosophila exhibits inter- and intra
specific variation in several factors, including auditory and chemosensory components,
mating speed, and courtship vigor. Certain aspects of courtship song, such as interpulse
interval, show species-specificity (inter-specific variation) thought to be an important
factor in reproductive isolation (Wheeler et al. 1991; Ritchie and Gleason 1995; Gleason
and Ritchie 2004). Gleason et al. (2002) identified chromosome regions containing
quantitative trait loci (QTL) for intra-specific variation in courtship song using a panel of
recombinant inbred lines derived from a cross between populations. These QTL had large
effects, but the regions did not contain genes previously suspected to affect courtship
behavior. Within population variation in courtship song also appears to have a genetic basis
in D. melanogaster, because it can respond to selection (Ritchie and Kyriacou 1994). This
does not always hold true for o th e r Drosophila species (Aspi and Hoikkala 1993).
Variation in female cuticular hydrocarbon (CHC) profiles can contribute to
species recognition and courtship interactions between males and females (Antony et al.
1985; Scott 1986; Schaner et al. 1989; Ferveur and Jallon 1996; Ferveur and Sureau
1996). Inter-specific (Coyne et al. 1994) and intra-specific variation in CHC profiles has
been reported (Tompkins and Hall 1984). The variation in CHC profiles parallels
variation in male response to CHCs both across and within species. To my knowledge
there have been no studies focused on intra-population variation in either female CHC
profile or variation in male response to differences in chemosensory cues.
Many studies of variation have focused on latency to copulation, or mating speed.
There is a great deal of inter-specific variation in mating speed due to differences in
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. female re-mating speed; if females of a species have high re-mating frequency, male
courtship behavior in that species is typically short and presumably not as important as
sperm competition (Markow 2002). There is also intra-specific variation in mating
speed; the trait responded to selection for both fast and slow mating speeds and genes
were identified as contributing to this phenotypic difference using both QTL (Moehring
and Mackay 2004) and micoarray analyses (Mackay et al. 2005). However, the majority
of the variation between these lines is due to female willingness to mate (Mackay et al.
2005). Casares and collegues verified the importance of female phenotype on variation
in mating speed within a population (Casares et al. 1993). However, other studies of
intra-population variation demonstrate significant effects of male phenotype
(Stamenkovicradak et al. 1992; Hoffmann 1999), suggesting that males may also have an
important role in mating speed.
Finally, there have been documented cases of within-population variation in
general courtship vigor (Gromko 1987), total time spent courting and performing
courtship song (Boake and Konigsberg 1998), and copulation duration (Boake and
Konigsberg 1998). None of the inter-specific, intra-specific, and intra-population studies
of variation in male courtship has examined the entire behavioral repertoire, or the pattern
of male behavior over time. Additionally, although genes that have effects on courtship
have been identified in mutant screening, it is not known whether these genes or an
entirely new set of genes are contributing to naturally occurring variation in behavior
(Hall 1994b; Greenspan 1997). Genes identified using mutant screening fall into several
categories based on mutant phenotype: general courtship vigor/success; courtship song;
olfaction; sex-determination; and learning/memory.
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Components of behavior and candidate genes
Several genes have been identified as playing a role in overall courtship drive and
vigor, couch potato (cpo) is one of several genes whose mutation causes a general
reduction in male courtship behavior, but does not disrupt one specific aspect of male
courtship, cpo is located on chromosome III and encodes a putative RNA recognition
motif protein, which is an RNA binding protein involved in the regulation of gene
expression at the post-transcriptional level (Lasko 2000). Mutations in cpo have revealed
that it is necessary for normal embryonic development of the peripheral nervous system
(PNS), and that it is also expressed in the central nervous system (CNS), PNS, and
salivary glands of adults (Bellen et al. 1992). It is also required for normal adult
behavior, as hypomorphic mutants have slow anesthesia recovery, decreased phototactic
and geotactic responses, and decreased flight ability—the flies are physically capable of
doing these behaviors, but are sluggish (Bellen et al. 1992). Mutants for this gene also
have severely reduced male courtship indices (Hall 1994a). cpo is an especially
intriguing candidate for this study because although not much is known about its specific
effects on male courtship, hypomorphic expression of this gene could have detrimental
effects on the success of a male in a natural population. Differential expression for cpo is
found in other behavioral traits, mating speed (Mackay et al. 2005) and male aggression
(Edwards et al. 2006), which indicates that its expression could be an important factor in
other behaviors such as male courtship.
Mutations in a locus on the X chromosome, technical knockout (tko), also have an
effect on the success of a male’s courtship, tko mutant males do not usually succeed in
initiating copulation with wild-type females, and when placed with tko mutant females,
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the number of successful copulations is greatly reduced compared to wild-type males
(Toivonen et al. 2001). Additionally, when tko mutant males are placed with tko mutant
females the latency to copulation is significantly longer than that of wild-type males,
indicating impairment in the general courtship ability of tko males (Toivonen et al. 2001).
tko encodes a ribosomal protein (Royden et al. 1987; Shah et al. 1997) that is part of the
redox reactions in the mitochondria; mutants have greatly decreased levels of redox
reactions, which are needed for normal energy output (Toivonen et al. 2001). tko is also
part of a class of “bang-sensitive” mutants that result in temporary paralysis after a
physical jolt (such as light vortexing) (Ganetzky and Wu 1982; Engel and Wu 1994;
Pavlidis and Tanouye 1995). This bang-sensitivity is due to a combination of hyper- and
hypoactivity in the sensory responses of the fly. Additionally, there is some indication
that thetko mutants are hearing impaired (Toivonen et al. 2001). Differences in the
expression of tko could influence behavior, because significant differential expression
was present between males in lines selected for fast and slow mating speed (Mackay et al.
2005).
Courtship song
Auditory cues are arguably the most important sensory aspect of courtship for
females in many species of Drosophila (Markow 1987). The male courtship song has
been implicated in species recognition (Wheeler et al. 1991). In D. montana, male
courtship song frequency correlates with the fitness of his progeny, and may be used as
an indicator of “good genes” by the female (Hoikkala et al. 1998). The song consists of
several distinct parts, including a series of pulses and a sinusoidal humming ‘sine song’
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (von Schilcher 1976). The overall mean interval between pulses (interpulse interval, IPI)
is species specific, as are the IPI oscillation periods (IPIs vary slightly between pulse
songs, and oscillate sinusoidally; (Kyriacou and Hall 1982)). For D. melanogaster, the
overall mean IPI is 34 milliseconds and the IPI oscillation period is 55 seconds (Kyriacou
and Hall 1982). Prior exposure to the pulse song of the appropriate species enhances
female receptivity to male courtship (Kyriacou and Hall 1984), and both the pulse and
sine songs decrease female locomotor activity to promote copulation (von Schilcher
1976).
Mutations in the atonal (ato) locus on chromosome III result in abnormal
courtship song production. Originally isolated as a developmental mutation, ato encodes
a basic helix-loop-helix protein involved in DNA binding (Jarman et al. 1995). ato is the
proneural gene responsible for the proper development of the chordotonal organs and
photoreceptors (Jarman et al. 1994; Jarman et al. 1995). Mutants for ato lack Johnston’s
Organ, which is responsible for hearing (Eberl et al. 2000). Without the ability to hear,
the courtship song produced by atonal mutants is significantly different than that of
wildtype flies, possibly due to the lack of sensory feedback mechanisms available to
normal individuals (Tauber and Eberl 2001). ato is expressed in adult male heads (Loop
et al. 2004), and thus the product must be used after development. Subtle differences in
the expression of this gene during adulthood could result in the production of an altered
courtship song in males, reducing their courtship intensity and lowering their fitness.
The dissonance allele of no-on-transient A (nonA
gene involved in the production of the male courtship song as well as influencing normal
vision (Kulkarni et al. 1988; Jones and Rubin 1990). nonA is on the X chromosome and
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. encodes an RNA binding protein that resembles a transcription factor and/or an RNA
splicing protein (Rendahl et al. 1992; Stanewsky et al. 1993; Rendahl et al. 1996). The
gene has two potential RNA recognition motifs (RRMs) that are common in RNA
binding proteins (Stanewsky et al. 1993; Rendahl et al. 1996). Only one of these RRMs
is important in controlling both the visual and song production phenotypes (Rendahl et al.
1996). nonAdlss mRNA is expressed in the CNS throughout development and continues
into adulthood. Turning on the expression of nonA during development, adulthood, or
both results in the rescue of normal visual and song production phenotypes, indicating
that thenonA protein product is used in the CNS throughout the entirety of the fly’s life
(Rendahl and Hall 1996).
nonAd,ss mutants have significantly lower courtship indices than wild-type males,
presumably due to both the visual and song irregularities. The male courtship song of
these mutants has a highly irregular pulse song phenotype—the pulse cycles in the
beginning of the pulse song resemble those of wildtype males, but as the song continues
the pulses become polycyclic and have irregular frequencies (Kulkarni et al. 1988). The
role of nonA in the production of courtship song has been supported by a study that
introduced the nonA transcript from D. virilis into nonAdlss mutant D. melanogaster males.
The D. virilis nonA transcript rescued the visual phenotype of the transformed D.
melanogaster, and resulted in the production of a courtship song that closely resembled
that of wild-type D. virilis (Campesan et al. 2001). This implicates nonA as an important
component of the species-specificity of male courtship song. Any differences in the
expression of this gene could result in an irregular courtship song, which could
dramatically affect the successful courtship of the male. Expression differences have
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. already been detected in nonA for lines of flies selected for fast vs. slow mating speed
(Mackay et al. 2005) or high vs. low male aggression (Edwards et al. 2006). These
selection regimes both began with base populations collected from the wild; hence, nonA
could have differential expression in natural populations.
Olfactory cues
Chemosensory cues are used in male assessment of the receptivity of a potential
partner, and olfactory cues have been well described. The courtship stimulating
compounds of mature females were identified as cis cis 7,11 heptacosadiene and cis cis
7,11 nonacosadiene (Antony et al. 1985). Immature males have courtship stimulating
alkenes, cis 13 tritriacontene and cis 11 tritriacontene (Schaner et al. 1989), that are not
counterbalanced by an inhibitory compound found on mature males, cis 7 tricosene (Scott
1986). The duration of male courtship depends on the balance of male inhibitory and
female stimulatory pheromones, and each of these seem to be independently processed
(Ferveur and Sureau 1996). Recent findings indicate that the female hydrocarbons are
used to modulate male responses to ancestral pheromones common to several species,
and that sexual isolation might occur in part because the males learn which modulatory
pheromones are associated with female acceptance of courtship attempts (Savarit et al.
1999).
One of the only known genes expressed in the brain that influences olfaction and
courtship behavior is paralytic (para). Certain mutations in para, smellblind (sbl) and
olfaction-deficient (olf-D), alter the responses of both larvae and adults to olfactory cues
(Tompkins et al. 1980; Tompkins et al. 1982; Gailey et al. 1986; Lilly and Carlson 1990);
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. these were later mapped to the same region on the X chromosome as para, which had
originally been isolated due to mutants displaying reversible paralysis at high
temperatures (Lilly et al. 1994). para encodes a voltage-sensitive sodium channel protein
a subunit, which is thought to play an important role in mediating neural membrane
excitability (Loughney et al. 1989). para is alternatively spliced, with mRNA from
several transcripts detected in the CNS (Loughney et al. 1989; Amichot et al. 1993).
These alternatively spliced regions are highly conserved across D. melanogaster and D.
virilis, indicating that para proteins are important for normal function in the fly
(Thackeray and Ganetzky 1995). Thesbl and olf-D alleles exhibit deficiencies in
olfaction, while other para alleles do not; one possible explanation for this is that sbl is
an alternative splice variant of para, and that this sodium-channel variant is important for
olfaction (Lilly et al. 1994).
sbl and olf-D mutant males have significantly reduced courtship indices when
paired with virgin females and immature males, but have significantly increased
courtship indices when presented with mated females (Tompkins et al. 1980; Gailey et al.
1986). Their inability to detect cuticular hydrocarbon profiles impairs their ability to
court well, and their orientation and courtship pattern is often incorrect when compared to
wild-type males (Tompkins et al. 1980; Markow 1987). Decreased expression of this
particular alternative splice variant in adults could significantly influence male courtship
intensity, as any variability in the number or quality of this sodium-channel a subunit
could have an effect on olfactory capabilities.
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sex-determination pathway
TheDrosophila sex-determination cascade involves many genes that are
alternatively spliced and expressed in a sex-specific manner, beginning with the
determination of the X: autosomal chromosome ratio during development. The reading of
the X:A ratio begins a male- or female-specific pathway—if thesex-lethal gene is turned
on by the genes responding to the X:A ratio, the female pathway has been chosen. The
transcription factor produced by sex-lethal then turns on female-specific splicing of the
transformer (tra) and transformer-2 (tra-2) genes, while simultaneously preventing
expression of male-specific fruitless splice variants. The tra gene then prevents the
expression of the male-specific form of doublesex. The male pathway is thus the default;
if sex-lethal is not expressed, tra and tra-2 are not activated, and the male-specific forms
of doublesex and fruitless are turned on, resulting in male behavior and appearance
(reviewed in MacDougall et al. 1995; Wolfner 2003).
Thefruitless gene on chromosome III is involved in the sex-determination
cascade. Mutations in fruitless (fru) result in many different phenotypes, ranging from
bisexual orientation to a complete abolition of courtship behavior (Gailey and Hall 1989).
Some fru mutants also have abnormal courtship songs (Villella et al. 1997).
Additionally, it seems to establish sex-specific aggressive behavior patterns and
dominance relationships (Vrontou et al. 2006). fru is alternatively spliced under the
control of tra and tra-2 (Heinrichs et al. 1998), and yields sex-specific transcripts; the
male-specific form of fruitless encodes a BTB zinc-finger protein, which is a putative
transcription factor, indicating that fru is involved in the regulation of a gene or genes
further down the sex-determination cascade (Ito et al. 1996; Ryner et al. \996). fru is also
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. expressed in the central nervous system of adult male flies, suggesting that it is involved
in gene regulation after development is complete (Goodwin et al. 2000; Lee et al. 2000).
The expression of the male-specific FRU protein also coincides with the presence of
octopamine in several locations in the male CNS, and fru seems to have implications for
controlling the switch from performing aggressive and courtship behaviors (Certel et al.
2007). Additionally, yh/ has highly variable expression amongst adult D. melanogaster
lines isogenic for the third chromosome (Hughes et al. 2006), and exhibits differential
expression in lines selected for fast and slow mating behavior (Mackay et al. 2005).
doublesex (dsx ) is also located on the third chromosome and is involved in the late
stages of the sex-determination cascade. Dependent on the presence of tra and tra-2, the
dsx gene can be alternatively spliced—the use of the default splice site results in male-
specific dsx expression, while tra and tra-2 allow the splicosome to use an alternate
splice-site, resulting in the expression of the female-specific dsx product (Burtis and
Baker 1989). Both dsx products encode transcription factors; DSXM actively suppresses
the expression of female-specific products such as yolk proteins (Coschigano and
Wensink 1993; An and Wensink 1995), while simultaneously activating the expression of
male-specific characteristics such as sex-comb development (Jursnich and Burtis 1993).
Peak dsx expression occurs during mid-pupal development, but expression continues
throughout adulthood in the CNS and the thoracic ganglia (Lee et al. 2002). This
expression is also sexually dimorphic, with males expressing much higher levels of dsx
during adulthood (Lee et al. 2002). There may be natural variation in expression of dsx,
as expression alters in response to selection for both mating speed (Mackay et al. 2005)
and male aggression (Edwards et al. 2006).
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Learning and memory
Male courtship behavior is modified by experience and environment. Males who
have been housed with other males (and exposed to homosexual courtship) achieve
successful copulation with females more quickly than naive males (Mcrobert and
Tompkins 1988). Males also learn to associate cis 7 tricosene (produced both by mature
males and mated females) with rejection behavior (Tompkins and Hall 1984), and modify
their courtship behavior in response to this association. This modification has been
termed “experience-dependent courtship conditioning”: males previously placed with
mated females show depressed mating activity (Siegel and Hall 1979). This conditioning
represses male courtship towards virgin females for 3-4 hours after contact with an
unreceptive female (Siegel and Hall 1979), but males retain the aversion to mated
females for approximately 24 hours (Gailey et al. 1984), supporting the idea that
courtship conditioning may exist to help males conserve energy and avoid courting
unreceptive females. However, the repression of courtship towards virgin females is
obviously not beneficial to the conditioned male. There does seem to be an adaptive
benefit of courtship conditioning conferred to males raised in an environment where they
encounter females in various states of receptivity. Males kept in constant contact with
many other flies for 21 generations showed enhanced courtship conditioning relative to
males with little exposure to females for the same number of generations (Reif et al.
2002).
Experience-dependent courtship conditioning has been identified as a form of
associative learning (Gailey et al. 1984), and the plasticity of learning and memory plays
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. an important role in the retention of this conditioning (Griffith et al. 1993). The gene
encoding calcium!calmodulin-dependent protein kinase II (CaMKlt) on chromosome IV
was first identified due to its functional homology to rat CaM kinase, which has been
implicated in learning (Griffith et al. 1993). After placement with a mated female for an
hour, males expressing an inhibitor of CaMKII protein do not retain courtship
conditioning—the time spent courting a virgin female after conditioning is significantly
greater than that of wild-type males who have been through the same training (Griffith et
al. 1993). The level of expression of CaMKII product has a direct effect on the intensity
of the resistance to conditioning (Griffith et al. 1993). CaMKII product potentially
phosphorylates the Ether-a-gogo (EAG) potassium-channel subunit, which can modulate
and regulate the post-synaptic potential of a repeatedly stimulated neuron by allowing
proper repolarization of the nerve and regulation of neurotransmitter release (Griffith et
al. 1994). Learning itself seems to occur due to an increase in postsynaptic potential
(modulated by potassium current) in neurons during sensitization (Schwartz and
Greenberg 1987).
Mutations at the X-chromosome locus dunce (dnc ) result in males deficient in
normal learning, including experience-dependent courtship conditioning (Byers et al.
1981; Ackerman and Siegel 1986; Dauwalder and Davis 1995). dnc encodes an
alternatively spliced cyclic-AMP phosphodiesterase (cAMP PDE), which helps
metabolize cAMP (Chen et al. 1986; Qiu et al. 1991; Qiu and Davis 1993). Experiments
indicate that cA M P must be present in the mushroom bodies of the fly in order for normal
associative and nonassociative learning to occur. There is speculation that without
certain dnc products available to keep cAMP to a minimum when it is not needed, the
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. presence of cAMP during training stages fails to properly modulate the neurons used in
learning (Davis and Dauwalder 1991). This can be detrimental to the mating success of a
male in a large population, as wild-type males who have learned to avoid courting
immature males have higher courtship indices and shorter latency to copulation with
virgin females than dnc mutants (Gailey et al. 1985).
Conclusion
I Use of natural variation to investigate the relationship between genes and
behavior has been advocated recently (Greenspan 1995; Sokolowski 2001; Ben-Shahar et
al. 2002; Emmons and Lipton 2003; Fitzpatrick et al. 2005). The approach used in my
dissertation integrates molecular techniques and detailed examination of naturally
occurring behavior to investigate within-population variation in behavior. Although
within-population variation is the raw material for evolution, the identity and nature of
contributing genes are largely unknown. With my dissertation, I have begun to dissect
genetic and molecular mechanisms that underlie natural variation in a complex
phenotype. The objectives of my research are included below.
Most of the studies of natural variation in courtship within populations of D.
melanogaster have focused on one component of behavior (i.e. song production).
However, valuable information can be lost when the details of the entire courtship
progression are not examined, because individual animals have subtle differences in their
patterns of behavior over time (Markow and Hanson 1981).
Objective 1: Develop a high-throughput assay of male courtship behavior that captures the details of courtship progression for a sample size large enough for genetic and
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. genomic studies. Use this male courtship progression (MCP) assay to determine the presence of genetic variation in the whole sequence of courtship behavior within a natural population.
After the presence of variation in MCP within a natural population was detected, I
then wanted to investigate the possible molecular contributions to natural variation. I
focused on differential gene expression, as alteration of the expression of a gene can be
one target of selection (Harris-Warrick 2000; Insel and Young 2000). Part of this
alteration can be due to non-genotypic factors, such as age or experience.
Objective 2: Examine courtship genes previously identified in the literature for alteration in gene expression due to age or experience with a female. Genes were considered static if age, experience, or the interaction between age and experience had no effect on expression. Genes were considered dynamically expressed if any of these factors had a significant effect on expression.
In the set of previously identified courtship genes, I found that any dynamic
changes in gene expression are correlated with age (reproductive maturity), not
experience with females. Thus, I continued my investigation of genes contributing to
natural variation in male courtship progression using mature males. I wanted to
determine which genes exhibited differential baseline expression between families with
the most extreme differences in courtship progression from Objective 1.
Objective 3: Use full-genome microarray analysis to find candidate genes associated with High and Low male courtship progression. Compare these candidates with
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. previously identified candidates, and examine functional trends in the new set of candidates.
I discovered 14 candidate genes; none were previously associated with male courtship
behavior. In future research, I will confirm that these candidates are directly associated
with male courtship progression by using functional mutants with decreased (or silenced)
expression to examine their effects on behavior.
List of References
Ackerman SL, Siegel RW (1986) Chemically reinforced conditioned courtship in drosophila - responses of wild-type and the dunce, amnesiac and don giovanni mutants. J Neurogenet 3(2): 111-123.
Amichot M, Castella C, Berge JB, Pauron D (1993) Transcription analysis of the para gene by insitu hybridization and immunological characterization of its expression product in wild-type and mutant strains of Drosophila. Insect Biochem Molec 23(3): 381-390. An WQ, Wensink PC (1995) Integrating sex-specific and tissue-specific regulation within a single Drosophila enhancer. Gene Dev 9(2): 256-266. Andersson MB (1994) Sexual Selection. Princeton, N. J.: Princeton University Press. 599 Antony C, Davis TL, Carlson DA, Pechine JM, Jallon JM (1985) Compared behavioral- responses of male Drosophila-melanogaster (canton-s) to natural and synthetic aphrodisiacs. J Chem Ecol 11(12): 1617-1629.
Aspi J, Hoikkala A (1993) Laboratory and natural heritabilities of male courtship song characters in Drosophila montana and D. Littoralis. Heredity 70(4): 400-406.
Bastock M, Manning A (1955) The courtship ofDrosophila melanogaster. Behaviour 8: 7-110. Bellen HJ, Vaessin H, Bier E, Kolodkin A, Develyn D et al. (1992) The Drosophila couch-potato gene - an essential gene required for normal adult behavior. Genetics 131(2): 365-375.
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ben-Shahar Y, Robichon A, Sokolowski MB, Robinson GE (2002) Influence of gene action across different time scales on behavior. Science 296(5568): 741-744.
Boake CRB, Konigsberg L (1998) Inheritance of male courtship behavior, aggressive success, and body size in Drosophila silvestris. Evolution 52(5): 1487-1492. Burtis KC, Baker BS (1989) Drosophila doublesex gene controls somatic sexual- differentiation by producing alternatively spliced messenger-RNAs encoding related sex-specific polypeptides. Cell 56(6): 997-1010. Byers D, Davis RL, Kiger JA (1981) Defect in cyclic-amp phosphodiesterase due to the dunce mutation of learning in Drosophila-melanogaster. Nature 289(5793): 79-81. Campesan S, Dubrova Y, Hall JC, Kyriacou CP (2001) The nonA gene in Drosophila
conveys species-specific behavioral characteristics. Genetics 158(4): 1535-1543. Casares P, Carracedo MC, Miguel ES, Pineiro R, Garciaflorez L (1993) Male mating speed in Drosophila-melanogaster - differences in genetic architecture and in relative performance according to female genotype. Behavior Genetics 23(4): 349-358. Certel SJ, Savella MG, Schlegel DCF, Kravitz EA (2007) Modulation of Drosophila male behavioral choice. Proceedings of the National Academy of Sciences of the United States of America 104(11): 4706-4711. Chen CN, Denome S, Davis RL (1986) Molecular analysis of cdna clones and the corresponding genomic coding sequences of the Drosophila dunce+ gene, the structural gene for cAMP phosphodiesterase. Proceedings of the National Academy of Sciences of the United States of America 83(24): 9313-9317. Coschigano KT, Wensink PC (1993) Sex-specific transcriptional regulation by the male and female doublesex proteins of Drosophila. Gene Dev 7(1): 42-54. Coyne JA, Crittenden AP, Mah K (1994) Genetics of a pheromonal difference contributing to reproductive isolation in Drosophila. Science 265(5177): 1461- 1464. Dauwalder B, Davis RL (1995) Conditional rescue of the dunce learning memory and female fertility defects withDrosophila or rat transgenes. Journal of Neuroscience 15(5): 3490-3499.
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Davis RL, Dauwalder B (1991) The Drosophila dunce locus - learning and memory genes in the fly. Trends Genet 7(7): 224-229. Dubnau J, Chiang AS, Grady L, Barditch J, Gossweiler S et al. (2003) The staufen/pumilio pathway is involved in Drosophila long-term memory. Current Biology 13(4): 286-296. Eberl DF, Hardy RW, Kernan MJ (2000) Genetically similar transduction mechanisms for touch and hearing in Drosophila. Journal of Neuroscience 20(16): 5981-5988. Edwards AC, Rollmann SM, Morgan TJ, Mackay TFC (2006) Quantitative genomics of aggressive behavior in Drosophila melanogaster. Plos Genetics 2(9): 1386-1395. Emmons SW, Lipton J (2003) Genetic basis of male sexual behavior. J Neurobiol 54(1): 93-110.
Engel JE, Wu CF (1994) Altered mechanoreceptor response in Drosophila bang-sensitive mutants. J Comp Physiol A 175(3): 267-278. Etges WJ (2002) Divergence in mate choice systems: Does evolution play by rules?
Genetica 116(2-3): 151-166. Ferveur JF, Sureau G (1996) Simultaneous influence on male courtship of stimulatory and inhibitory pheromones produced by live sex-mosaic Drosophila melanogaster. P Roy Soc Lond B Bio 263(1373): 967-973. Ferveur JF, Jallon JM (1996) Genetic control of male cuticular hydrocarbons in Drosophila melanogaster. Genetical Research 67(3): 211-218. Fitzpatrick M, Ben-Shahar Y, Smid H, Vet L, Robinson G et al. (2005) Candidate genes for behavioural ecology. Trends Ecol Evol 20(2): 96-104. Gailey DA, Hall JC (1989) Behavior and cytogenetics of fruitless in Drosophila-
melanogaster - different courtship defects caused by separate, closely linked lesions. Genetics 121(4): 773-785. Gailey DA, Jackson FR, Siegel RW (1984) Conditioning mutations in Drosophila- melanogaster affect an experience-dependent behavioral-modification in courting males. Genetics 106(4): 613-623. Gailey DA, Hall JC, Siegel RW (1985) Reduced reproductive success for a conditioning mutant in experimental populations of Drosophila-melanogaster. Genetics 111(4): 795-804.
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gailey DA, Lacaillade RC, Hall JC (1986) Chemosensory elements of courtship in normal and mutant, olfaction-deficient Drosophila-melanogaster. Behavior Genetics 16(3): 375-405. Ganetzky B, Wu CF (1982) Indirect suppression involving behavioral mutants with altered nerve excitability in Drosophila-melanogaster. Genetics 100(4): 597-614. Gleason JM, Ritchie MG (2004) Do quantitative trait loci (qtl) for a courtship song difference between Drosophila simulans and D. sechellia coincide with candidate genes and intraspecific QTL? Genetics 166(3): 1303-1311. Gleason JM, Nuzhdin SV, Ritchie MG (2002) Quantitative trait loci affecting a courtship signal in Drosophila melanogaster. Heredity 89: 1-6. Goodwin SF, Taylor BJ, Villella A, Foss M, Ryner LC et al. (2000) Aberrant splicing and altered spatial expression patterns in fruitless mutants of Drosophila melanogaster. Genetics 154(2): 725-745. Greenspan RJ (1995) Understanding the genetic construction of behavior. Sci Am 272(4):
72-78. Greenspan RJ (1997) A kinder, gentler genetic analysis of behavior: Dissection gives
way to modulation. Curr Opin Neurobiol 7(6): 805-811. Greenspan RJ, Ferveur JF (2000) Courtship in Drosophila. Annu Rev Genet 34: 205-232. Griffith LC, Wang J, Zhong Y, Wu CF, Greenspan RJ (1994) Calcium/calmodulin- dependent protein-kinase-II and potassium channel subunit eag similarly affect plasticity in Drosophila. Proceedings of the National Academy of Sciences of the United States of America 91(21): 10044-10048. Griffith LC, Verselis LM, Aitken KM, Kyriacou CP, Danho W et al. (1993) Inhibition of
calcium calmodulin-dependent protein-kinase in Drosophila disrupts behavioral plasticity. Neuron 10(3): 501-509. Gromko MH (1987) Genetic constraint on the evolution of courtship behavior in Drosophila-melanogaster. Heredity 58: 435-441. Hall JC (1994a) The mating of a fly. Science 264(5166): 1702-1714. Hall JC (1994b) Pleiotropy of Behavioral Genes. In: Greenspan RJ, Kyriacou CP, editors. Flexability and constraint in behavioral systems: John Wiley & Sons, LTD. pp. 15-27.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Harris-Warrick RM (2000) Ion channels and receptors: Molecular targets for behavioral evolution. J Comp Physiol A 186(7-8): 605-616. Heinrichs V, Ryner LC, Baker BS (1998) Regulation of sex-specific selection of fruitless 5 ' splice sites by transformer and transformer-2. Mol Cell Biol 18(1): 450-458. Hoffmann AA (1999) Is the heritability for courtship and mating speed in Drosophila (fruit fly) low? Heredity 82: 151-157. Hoikkala A, Aspi J, Suvanto L (1998) Male courtship song frequency as an indicator of male genetic quality in an insect species, Drosophila montana. P Roy Soc Lond B Bio 265(1395): 503-508. Hughes KA, Ayroles JF, Reedy MM, Drnevich JM, Rowe KC et al. (2006) Segregating variation in the transcriptome: Cis regulation and additivity of effects. Genetics
173(3): 1347-1364. Insel TR, Young LJ (2000) Neuropeptides and the evolution of social behavior. Curr Opin Neurobiol 10(6): 784-789. Ito H, Fujitani K, Usui K, ShimizuNishikawa K, Tanaka S et al. (1996) Sexual orientation in Drosophila is altered by thesatori mutation in the sex- determination gene fruitless that encodes a zinc finger protein with a btb domain. Proceedings of the National Academy o f Sciences of the United States of America 93(18): 9687-9692. Jarman AP, Sun Y, Jan LY, Jan YN (1995) Role of the proneural gene, atonal, in formation of Drosophila chordotonal organs and photoreceptors. Development 121(7): 2019-2030. Jarman AP, Grell EH, Ackerman L, Jan LY, Jan YN (1994) Atonal is the proneural gene for Drosophila photoreceptors. Nature 369(6479): 398-400. Jones KR, Rubin GM (1990) Molecular analysis of no-on-transient-A, a gene required for normal vision in Drosophila. Neuron 4(5): 711-723. Jursnich VA, Burtis KC (1993) A positive role in differentiation for the male doublesex protein of Drosophila. Dev Biol 155(1): 235-249. Kulathinal RJ, Singh RS (2004) The nature of genetic variation in sex and reproduction- related genes among sibling species of the Drosophila melanogaster complex. Genetica 120(1-3): 245-252.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kulkarni SJ, Steinlauf AF, Hall JC (1988) The dissonance mutant of courtship song in Drosophila-melanogaster - isolation, behavior and cytogenetics. Genetics 118(2): 267-285.
Kyriacou CP, Hall JC (1982) The function of courtship song rhythms in Drosophila. Animal Behaviour 30(AUG): 794-801. Kyriacou CP, Hall JC (1984) Learning and memory mutations impair acoustic priming of mating-behavior in Drosophila. Nature 308(5954): 62-65. Lasko P (2000) The Drosophila melanogaster genome: Translation factors and RNA binding proteins. J Cell Biol 150(2): F51-F56. Lee G, Hall JC, Park JH (2002) Doublesex gene expression in the central nervous system of Drosophila melanogaster. J Neurogenet 16(4): 229-248. Lee G, Foss M, Goodwin SF, Carlo T, Taylor BJ et al. (2000) Spatial, temporal, and sexually dimorphic expression patterns of the fruitless gene in the Drosophila central nervous system. J Neurobiol 43(4): 404-426. Lilly M, Carlson J (1990) Smellblind - a gene required for Drosophila olfaction. Genetics 124(2): 293-302. Lilly M, Kreber R, Ganetzky B, Carlson JR (1994) Evidence that theDrosophila olfactory mutant smellblind defines a novel class of sodium-channel mutation. Genetics 136(3): 1087-1096. Loop T, Leemans R, Stiefel U, Hermida L, Egger B et al. (2004) Transcriptional signature of an adult brain tumor in Drosophila. BMC Genomics 5: 24.
Loughney K, Kreber R, Ganetzky B (1989) Molecular analysis of the para locus, a sodium-channel gene in Drosophila. Cell 58(6): 1143-1154.
MacDougall C, Harbison D, Bownes M (1995) The developmental consequences of alternate splicing in sex determination and differentiation in Drosophila. Dev Biol 172(2): 353-376. Mackay TFC, Heinsohn SL, Lyman RF, Moehring AJ, Morgan TJ et al. (2005) Genetics and genomics of Drosophila mating behavior. Proceedings of the National Academy of Sciences of the United States of America 102: 6622-6629.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Markow TA (1987) Behavioral and sensory basis of courtship success in Drosophila- melanogaster. Proceedings of the National Academy of Sciences of the United States of America 84(17): 6200-6204. Markow TA (2002) Perspective: Female remating, operational sex ratio, and the arena of sexual selection in Drosophila species. Evolution 56(9): 1725-1734. Markow TA, Hanson SJ (1981) Multivariate analysis of Drosophila courtship. Proceedings of the National Academy of Sciences of the United States of America 78(1): 430-434. Mcrobert SP, Tompkins L (1988) 2 consequences of homosexual courtship performed by Drosophila-melanogaster and Drosophila-affinis males. Evolution 42(5): 1093-1097. Moehring A, Mackay T (2004) The quantitative genetic basis of male mating behavior in Drosophila melanogaster. Genetics 167(3): 1249-1263. Morozova TV, Anholt RRH, Mackay TFC (2006) Transcriptional response to alcohol exposure in Drosophila melanogaster. Genome Biology 7(10): 10. Pavlidis P, Tanouye MA (1995) Seizures and failures in the giant fiber pathway of
Drosophila bang-sensitive paralytic mutants. Journal of Neuroscience 15(8): 5810-5819. Qiu YH, Davis RL (1993) Genetic dissection of the learning memory gene dunce of Drosophila-melanogaster. Gene Dev 7(7B): 1447-1458. Qiu YH, Chen CN, Malone T, Richter L, Beckendorf SK et al. (1991) Characterization of the memory gene dunce of Drosophila-melanogaster. J Mol Biol 222(3): 553-565. Reif M, Linsenmair KE, Heisenberg M (2002) Evolutionary significance of courtship conditioning in Drosophila melanogaster. Animal Behaviour 63(Part 1): 143-155. Rendahl KG, Hall JC (1996) Temporally manipulated rescue of visual and courtship abnormalities caused by a nonA mutation in Drosophila. J Neurogenet 10(4): 247-256. Rendahl KG, Jones KR, Kulkarni SJ, Bagully SH, Hall JC (1992) The dissonance mutation at the no-on-transient-A locus of Drosophila-melanogaster - genetic- control of courtship song and visual behaviors by a protein with putative RNA- binding motifs. Journal of Neuroscience 12(2): 390-407.
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rendahl KG, Gaukhshteyn N, Wheeler DA, Fry TA, Hall JC (1996) Defects in courtship and vision caused by amino acid substitutions in a putative RNA-binding protein encoded by the no-on-transient A (nonA ) gene of Drosophila. Journal of Neuroscience 16(4): 1511-1522. Ritchie MG, Kyriacou CP (1994) Genetic-variability of courtship song in a population of Drosophila-melanogaster. Animal Behaviour 48(2): 425-434.
Ritchie MG, Gleason JM (1995) Rapid evolution of courtship song pattern in Drosophila-willistoni sibling species. J Evolution Biol 8(4): 463-479. Royden CS, Pirrotta V, Jan LY (1987) The tko locus, site of a behavioral mutation in Drosophila-melanogaster, codes for a protein homologous to prokaryotic ribosomal-protein sl2. Cell 51(2): 165-173. Ryner LC, Goodwin SF, Castrillon DH, Anand A, Villella A et al. (1996) Control of male sexual behavior and sexual orientation in Drosophila by thefruitless gene. Cell 87(6): 1079-1089. Savarit F, Sureau G, Cobb M, Ferveur JF (1999) Genetic elimination of known pheromones reveals the fundamental chemical bases of mating and isolation in Drosophila. Proceedings of the National Academy of Sciences of the United States of America 96(16): 9015-9020.
Sawamura K, Tomaru M (2002) Biology of reproductive isolation in Drosophila: Toward a better understanding of speciation. Population Ecology 44(3): 209-219. Schaner AM, Dixon PD, Graham KJ, Jackson LL (1989) Components of the courtship- stimulating pheromone blend of young male Drosophila-melanogaster - (z)-13- tritriacontene and (z)-ll-tritriacontene. J Insect Physiol 35(4): 341-345. Schwartz JH, Greenberg SM (1987) Molecular mechanisms for memory - 2nd-messenger induced modifications of protein-kinases in nerve-cells. Annu Rev Neurosci 10: 459-476. Scott D (1986) Sexual mimicry regulates the attractiveness of mated Drosophila- melanogaster females. Proceedings of the National Academy of Sciences of the United States of America 83(21): 8429-8433. Shah ZH, O'Dell KMC, Miller SMC, An X, Jacobs HT (1997) Metazoan nuclear genes for mitoribosomal protein sl2. Gene 204(1-2): 55-62.
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Siegel RW, Hall JC (1979) Conditioned responses in courtship behavior of normal and mutant Drosophila, Proceedings of the National Academy of Sciences of the United States of America 76(7): 3430-3434.
Sokolowski MB (2001) Drosophila'. Genetics meets behaviour. Nat Rev Genet 2(11): 879-890. Sorensen JG, Nielsen MM, Kruhoffer M, Justesen J, Loeschcke V (2005) Full genome gene expression analysis of the heat stress response, in Drosophila melanogaster. Cell Stress & Chaperones 10(4): 312-328. Stamenkovicradak M, Partridge L, Andjelkovic M (1992) A genetic correlation between the sexes for mating speed in Drosophila-melanogaster. Animal Behaviour 43(3):
389-396. Stanewsky R, Rendahl KG, Dill M, Saumweber H (1993) Genetic and molecular analysis of the x-chromosomal region 14bl7-14c4 in Drosophila-melanogaster - loss of function in nonA, a nuclear-protein common to many cell-types, results in specific physiological and behavioral defects. Genetics 135(2): 419-442. Stirling DG, Reale D, Roff DA (2002) Selection, structure and the heritability of behaviour. J Evolution Biol 15: 277-289. Tauber E, Eberl DF (2001) Song production in auditory mutants of Drosophila : The role of sensory feedback. J Comp Physiol A 187(5): 341-348. Thackeray JR, Ganetzky B (1995) Conserved alternative splicing patterns and splicing signals in the Drosophila sodium-channel gene para. Genetics 141(1): 203-214. Toivonen JM, O'Dell KMC, Petit N, Irvine SC, Knight GK et al. (2001) technical knockout, a Drosophila model of mitochondrial deafness. Genetics 159(1): 241-254. Tompkins L, Hall JC (1984) Sex-pheromones enable Drosophila males to discriminate between conspecific females from different laboratory stocks. Animal Behaviour 32(MAY): 349-352. Tompkins L, Hall JC, Hall LM (1980) Courtship-stimulating volatile compounds from normal and mutant Drosophila. J Insect Physiol 26: 689-697.
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tompkins L, Gross AC, Hall JC, Gailey DA, Siegel RW (1982) The role of female movement in the sexual-behavior of Drosophila-melanogaster. Behavior Genetics 12(3): 295-307. Villella A, Gailey DA, Berwald B, Ohshima S, Barnes PT et al. (1997) Extended reproductive roles of the fruitless gene in Drosophila melanogaster revealed by behavioral analysis of new fru mutants. Genetics 147(3): 1107-1130.
von Schilcher F (1976) The role of auditory stimuli in the courtship of Drosophila melanogaster. Animal Behavior 24: 18-26. Vrontou E, Nilsen SP, Demir E, Kravitz EA, Dickson BJ (2006) fruitless regulates aggression and dominance in Drosophila. Nature Neuroscience 9(12): 1469-1471. Wheeler DA, Kyriacou CP, Greenacre ML, Yu Q, Rutila JE et al. (1991) Molecular transfer of a species-specific behavior from Drosophila-simulans to Drosophila- melanogaster. Science 251(4997): 1082-1085.
Wolfner MF (2003) Sex determination: Sex on the brain? Current Biology 13(3): R101-R103.
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 2: Natural variation in the courtship progression of
male Drosophila melanogaster
Co-authored by Kimberly A. Hughes1
Abstract
Courtship behavior is a main component of male fitness, especially in species
with no parental care. Variation in this behavior can thus be a target for mate choice and
sexual selection, and can lead to evolution. The fruit fly, Drosophila melanogaster, has
well-documented complex male courtship comprised of a progressive sequence of
distinct behaviors, and is an ideal model for behavior-genetic analysis. In order to
evaluate genetic differences in the temporal pattern of behavior, I have developed a
simple, high-throughput method that allows the documentation of the progression of male
courtship using an ordinal scale (male courtship progression scale, MCP). Using this
method, I document natural genetic variation in the temporal pattern of behavior that was
not detected using other metrics. This method was robust enough to detect genetic
variation in this trait for males placed with both virgin and mated female targets.
Introduction
Courtship behavior can be a key determinant of male fitness in animal species that
have little or no male parental care. It can signal species identity and can be the object of
intersexual selection via direct benefits, indirect benefits, or sensory bias modalities
(Andersson 1994). Despite potentially strong directional selection, which should erode
1 Ruedi, E.A. and Hughes, K.A. In review. Natural variation in the courtship progression o f male Drosophila melanogaster. Submitted to Behavior Genetics.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. variation, courtship behavior exhibits substantial genetic variation within populations of
many species (Stirling et al. 2002). Because variation in courtship behavior can lead to
reproductive isolation and speciation, evolutionary effects of such variation are
particularly interesting (Etges 2002; Sawamura and Tomaru 2002; Kulathinal and Singh
2004). To successfully understand the causes and consequences of this variation, we
must 1) identify the presence of genetic differences in a natural population, 2) investigate
the molecular mechanisms underlying variation, and 3) identify the evolutionary
processes responsible for maintenance of variation. In this study I address the first of
these points. I also present a high throughput method for obtaining robust, detailed
measures of courtship behavior and compare this method with others currently available.
Drosophila melanogaster is a model system for genetic analysis of behavior. Flies
in the genus Drosophila have complex male courtship that is characterized by a
progressive sequence of distinct behaviors that culminates in copulation (Hall 1994). In
D. melanogaster males, the courtship progression includes orienting toward the female,
following her, tapping her abdomen, performing a courtship “song” via wing vibration,
licking her genitalia, attempting copulation, and copulation (Hall 1994). This behavioral
progression involves visual, auditory and chemical communication between males and
females, as well as aspects of learning and memory (Hall 1994; Yamamoto and Nakano
1998; Greenspan and Ferveur 2000).
Variation in male courtship has been demonstrated between closely related
species ofD rosoph ila (Ritchie and Gleason 1995; Markow etal. 2002; Gleason and
Ritchie 2004), between populations within species (ex. Gleason and Ritchie 2004;
Moehring and Mackay 2004; Mackay et al. 2005) and within populations (ex.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stamenkovicradak et al. 1992; Aspi and Hoikkala 1993; Ritchie and Kyriacou 1994;
Boake and Konigsberg 1998; Hoffmann 1999). However, many of these studies have
focused on one or a few specific behaviors in the courtship repertoire. Valuable
information can be lost when the details of the entire courtship progression are not
examined, because individual animals have subtle differences in their patterns of
behavior over time (Markow and Hanson 1981). Also, most previous studies have
assayed only male behavior towards virgin females. It is unknown if variation in
courtship towards virgin females is representative of courtship towards mated females.
In nature and in most laboratory culture conditions, most encounters between sexes will
involve previously-mated female targets (Bouletreau 1978; Partridge et al. 1987). Thus
the most evolutionarily relevant variation in courtship behavior may be that between
males when they encounter non-virgin females (Gromko and Markow 1993; Reif et al.
2002).
Most analyses of fly courtship have used metrics that either focus on a specific
behavior in the sequence or use a method that does not discriminate between different
behaviors in the sequence. For example, the commonly-used Courtship Index (Cl)
measures the proportion of the total observational trial that the male spends performing
any courtship behavior (eg. Siegel and Hall 1979). Other metrics include latency to
copulation (Moehring and Mackay 2004; Mackay et al. 2005), and copulation occurrence
and duration (e.g. Beaver and Giebultowicz 2004; Moehring and Mackay 2004). A
notable exceptionis the approach of Markow and Hanson(1981), w ho used detailed
analysis of focal animal samples to characterize temporal patterns of behavior. A more
detailed metric is more sensitive and more likely to pick up small, but critically important
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. behavioral variation that has a genetic basis; however, the focal animal approach is labor
intensive and difficult to apply to the large sample size needed in genetic and genomic
experiments. Therefore, we need to refine behavioral methods to compare differences
between individuals and keep pace with the genetic and genomic methods that require
high throughput phenotypic analysis.
Here, I developed a novel approach for quantifying the patterns of male courtship
that is sensitive to differences in both overall courtship activity and to temporal patterns
in activity and can be used for high throughput phenotypic analysis. I developed an
ordinal scale that mirrors the normal progression of courtship in D. melanogaster males,
and applied this method to analysis of courtship of males placed either with virgin or
mated females. I detected significant genetic differences between families from a single
wild-type population for males exposed to both kinds of females. Using this method, I
can quantify the extent and structure of genetic variation in courtship behavior and
examine differences resulting from the mating history of the female target.
Methods
Courtship Assay
D. melanogaster courtship typically follows the temporal sequence specified in
Table 2.1, culminating in successful copulation. I therefore assigned each behavior to an
ordinal value ranging from 1 (motionless) to 8 (copulation). The ordinal nature of this
male courtship progression (MCP) scale implies a qualitative measure of distance
between behaviors (e.g., licking is “closer” to copulation than is orienting) without
specifying a quantitative measure of distance between the behaviors (Martin and Bateson
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1993). This ordinal scale, combined with repeated scan sampling during behavior
observations, can be applied to many mating trials simultaneously. When used with
appropriate statistical analysis, the technique can be used to quantify the intensity,
duration, and temporal pattern of a complex behavior, and can be efficiently used in
large-scale experiments requiring the observation of many individuals.
Experimental Flies
I used flies descended from wild-caught ancestors collected from a site in
Terhune, New Jersey by Valerie Pierce. The founding population consisted of
approximately 8,000 offspring from 400 females collected in 1999. This population was
maintained in the laboratory of Dr. Allen Gibbs at a population size of 1000-1500 for 45
generations (A. Gibbs, pers. comm.) until the Hughes lab obtained ~ 500 individuals in
2003. Since then, we have maintained the population with overlapping generations at a
size of -12,000 individuals. This population (subsequently, the NJ population) is kept on
standard commeal media in 8-dram vials, transferred to new media every 14 days, and
housed at 25°C on a 12h:12h light:dark cycle. Flies are randomly mixed in bottles and
redistributed to vials every 28 days to maintain random mating.
I chose the NJ population for this study because it has been in the lab long enough
(>100 generations) that novel-environment effects will not substantially bias estimates of
genetic variation (Service and Rose 1985), but not so long that variation has been
dramatically eroded. Indeed,the population has responded vigorously to selection for
desiccation resistance (Gefen et al. 2006) and lifespan (K. Hughes, unpublished data),
indicating that it harbors substantial genetic variability.
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. To detect genetic differences in courtship, I used a full-sib mating design, using
160 different full-sib families in two different experiments (80 families for each
experiment). Because of the large number of families in each experiment, families were
assayed in four blocks of 20 families each for each experiment. Parents of experimental
families were reared from controlled-density vials, collected as virgins 0-4 hours post-
eclosion, and housed in single-sex vials for 5-7 days. I produced single-pair matings by
randomly pairing males and females from this collection, and introducing each pair to a
fresh rearing vial. Pairs were allowed to produce offspring for four days, transferred to a
new vial for four more days, and were then discarded. Collecting males from these two
different rearing vials reduced the possibility of confounding environmental effects on
behavior with the effect of family. I collected 5-10 virgin male offspring from both vials
for each single-pair mating at 0 to 4 hours post-eclosion over a three-day period, and the
males were kept in single-sex vials separated by family and by date of emergence. These
virgin males comprised my full-sib families. I used 80 of these full-sib families for
courtship assays in which virgin females were the mating targets (virgin target, or VT
experiment); I used another 80 families for courtship assays in which previously-mated
females were the mating targets (mated target, or MT experiment). Up to eight males per
family were used in behavior trials (mean= 7, range = 5-8). The VT and MT experiments
were performed at different times with different females.
Females used as mating targets were collected from a stock in which a
spontaneously-occurringebony (e) mutation had been introgressed into a non-inbred wild
type genetic background (Ives background), ebony females were used because of their
dark body color, which aids in observer discrimination between male and female flies in
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the behavioral trials. This stock, which is homozygous for thee mutation, is maintained
in a large, randomly-mating population with overlapping generations and reared under
the same conditions as the NJ population. The e/e females used in the behavioral trials
were collected as virgins 0-4 hours post-eclosion and kept in single-sex vials separated by
date of emergence. For each block of an experiment, 180 virgin females were collected
over a three-day period. For the VT experiment, the females were kept in their original
collection vials, at a density of 25 flies per vial, until the behavioral trial. Females from
the same vial were assigned to different male families in the behavior trial to avoid
confounding environmental effects shared among females with genetic effects of male
families. For the MT experiment, 180 virgin females were collected over a three day
period. 24 hours prior to the beginning of behavioral trials, subsets of 45 females were
placed in a half-liter bottle with media and with 50 e/e males for two hours, and then
housed separately from males in vials at a density of 25 females per vial. To ensure that
all females for the MT experiment had mated successfully with e/e males, any females
not observed in copula after twenty minutes of exposure to e/e males were removed via
aspiration and discarded.
Behavioral Observations
Behavior observations were conducted in devices modeled after Drapeau and
Long (2000), see Figure 2.1. Behavioral observations were carried out between July and
Decem ber, 2004. For each block of the VT and MT experiments, twomales from each
family were observed per day, and there were four days of observation per block. No
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. male was observed more than once, and observations were blind with respect to
genotype. All males were 5-6 days old when behavioral observations were conducted.
On the day before their trial, males were randomly assigned to one of forty arenas
(four separate devices with ten chambers each were used each day). Males were lightly
anesthetized with CO2 , and then placed in a chamber that comprised the bottom half of
each arena. After all males were added, the upper chambers were placed onto the
apparatus over opaque divider sheets. Females were lightly anesthetized and quickly
transferred over to the upper chambers (1 female/chamber). Finally, the clear plexiglass
top was attached, and the flies were allowed to recover from anesthesia overnight in an
incubator at 25°C.
All observations took place between 8 am and 10:30 am and were conducted at
24°C ± 1 °C, in a room isolated from all other activities. At the start of observations, the
opaque plastic sheet separating the sexes was removed; thus a single male was exposed to
a single female. The behavior of each male was then recorded via instantaneous scan
sampling (Martin and Bateson 1993) every 30 seconds for a total of 30 minutes. At each
scan, the male was scored as performing one of the behaviors in Table 2.1. At the end of
a trial, each male was thus characterized by 60 scores in a temporal sequence. All
observations were conducted by the same observer (E.A.R.). With this scheme, a total of
40 males could be observed per day in a two-hour time span (4 Vi-hour trials per day),
and each block consisted of four days of observation (160 males per block).
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Statistical Analyses
Families in the VT experiment were analyzed separately from families in the MT
experiment. All statistical analyses were conduced using SAS Statistical Analysis
Software v. 9.1 (SAS Institute 2006). Unless otherwise stated, dependent variables met
the assumptions of general linear models (normally and homoscedastically distributed).
Block effects and interactions of blocks with other factors were not significant in any
analyses; therefore, I report analyses with those effects removed.
In this experiment, I limited our attention to variation in male behavior, and I
tested males from each family using different, unrelated females. However, differences
among females can contribute to differences in male mating behavior (Reif et al. 2002). I
therefore interpret differences in male behavior between families as differences among
male genotypes, averaged over a random sample of female genotypes. In this experiment,
interaction between male and female genotype contributes to the residual variation in the
statistical analysis, and is thus not confounded with familial differences in male behavior.
Male Courtship Progression
The eight-point ordinal scale used to characterize male behavior resulted in a
multinomially distributed variable, with repeated measures on individuals at each 30-
second time interval. To analyze temporal patterns and family-by-time interactions for
this variable, I used a generalized linear model modified for repeated measures. This
model is a non-parametric alternative to ANOVA, and the generalized linear model
procedure SAS PROC GENMOD implements Generalized Estimating Equations (GEE)
to evaluate repeated-measures data structures (Littell et al. 2002). I used the cumulative
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. logit link function in PROC GENMOD to model the cumulative probabilities of the
ordinal MCP scores, and applied a repeated-measures analysis with individual flies as the
subject and the time-specific samples for each fly as the repeated measure. Only the first
30 time points of each trial (the first fifteen minutes) were used in the analysis, because
nearly all males had ceased locomotor activity after that time, and therefore had the same
MCP score: “motionless”. The model included both linear and quadratic effects of time
as predictor variables because preliminary inspection of the data indicated an
approximately quadratic change in MCP values over time (see Figure 2.2).
I first fit a model containing all terms and interactions of interest, and
subsequently removed interaction terms that were not significant. In the initial model,
the probability that an observation is a particular response out of the eight ordered
response categories of the MCP scale was modeled as: logit(Pj) = [i + f+ d + a(d) + t +
(t*t) + (f*t) +(f*t*t), where P, was the cumulative probability of the ordinal response i ,
and the other terms represent the fixed effects of family (f), day (d), trial (a) within day,
linear and quadratic effects of time (t) within the courtship trials (in 30-second intervals),
and interactions between family and time; fi, was the intercept for a particular ordinal
category. Because this is a repeated-measures analysis, a significant family effect
indicates that families are significantly different at time 0, and significant interactions
between family and time variables indicates that families have different temporal patterns
of behavior.
I was interested in detecting whether full-sib families differed from one another in
indices of courtship behavior, and also in identifying families with relatively high and
low levels of courtship behavior. This is a useful approach to identify families for studies
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the genetic basis of behavioral differences. Additionally, multinomial data is modeled
such that each observation has more than two possible outcomes (in this case, each
observation has 8 possible outcomes), and the expected mean for each observation is not
scalar. Because of this, variance is represented by a covariance matrix; thus it is not
possible to use the generalized linear model to calculate a single measure of “variance”
for MCP. For these reasons, I treated families as fixed effects in the analyses of MCP
and in derived metrics.
Comparing MCP to Other Metrics
I compared MCP to several commonly used metrics, including Cl, copulation
latency, occurrence, and duration, and a multivariate factor pattern analysis. Unless
otherwise noted, I analyzed these data using mixed linear models in SAS PROC MIXED.
The effects of family, day, and trial(day) were modeled as fixed effects. All univariate
metrics were compared by calculating the Spearman rank correlation between the metrics
across families. MCP can also be used to examine individual behaviors in the courtship
repertoire, as is seen in some literature (Gromko 1987; Gromko et al. 1991;
Stamenkovicradak et al. 1992; Aspi and Hoikkala 1993; Casares et al. 1993; Ritchie and
Kyriacou 1994; Boake and Konigsberg 1998). Results for this analysis are reported in
appendix A.
C l Proxy
I calculated a proxy for the most-commonly used composite measure of courtship
behavior, Cl, which is typically measured as the proportion of time during a trial that a
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. male spends performing any behavior related to courtship (eg. Siegel and Hall 1979;
Ferveur and Jallon 1993; Hall 1994; Ryner et al. 1996; Balakireva et al. 1998; Mackay et
al. 2005). Because each focal male was not observed continuously during his mating trial,
I could not calculate a true Cl from our data, but I could estimate Cl by dividing the
number of observations during which a male performed any of the MCP categories 3
through 8 (“orienting” through “copulation”) by the total number of observations for the
male. The dependent variable was transformed by taking the arcsine square root, as is
appropriate for proportions (Zar 1996).
Latency to copulation
“Mating speed” or “latency to copulation” has also been used as an index of
courtship behavior (Moehring and Mackay 2004; Mackay et al. 2005). I calculated an
equivalent metric for the VT experiment only, because very few families achieved
copulation in MT families. I recorded the latency to copulation as the observation period
that an individual reached the MCP score of 8 (copulation). In this analysis, all 60 time
periods of a mating trial were used to calculate latency. Any individuals that did not
achieve copulation during the trial were assigned a latency of “61”. Latency scores were
log transformed before analysis to improve fit to normality assumptions.
Copulation Duration and Occurrence
Some studies have evaluated mating behavior using measures of copulation
occurrence and duration (e.g. Beaver and Giebultowicz 2004; Moehring and Mackay
2004). To compare MCP to copulation occurrence, I tested for among-family differences
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in copulation occurrence using a generalized linear model (PROC GENMOD) with
binomially distributed error and logit link function. The model was: logit(jt) = p + f + d +
a(d), where it was the probability of copulation for an individual, and the predictor
variables are abbreviated as above. I tested for differences in copulation duration by
fitting the linear model described above to the proportion of observations during which
an individual was observed copulating (transformed with arcsine square root).
Multivariate Analysis: Factor Pattern
I performed a factor pattern analysis to compare my results to the multivariate
analysis of Drosophila courtship developed by Markow and Hanson (1981). This
analysis is used to detect if particular behaviors typically co-occur. Markow and Hanson
(1981) conducted a factor pattern analysis of male behavior data collected on 15 pairs of
courting flies, and then calculated transition matrices documenting behavioral changes
over the course of the trial. Replication of the transition matrix analysis is not possible
with my data because I did not continuously monitor each focal animal. However, I
could perform a maximum-likelihood (canonical) factor pattern analysis (PROC FACT),
and also obtain factor-loading scores for each male for analysis of genetic variation. The
model for this analysis was: yy= xubij + Xj2b2j + ... + xiqbqj + ey, where yy was the
proportion of time individual i spent performing behavior j, Xjk, k=i..q is the value of theith
observation on the k h common factor, bkj is the regression coefficient of the lth factor for
predicting the j ,h behavior, and ey is the value of the i"‘ observation loaded on a unique
factor for the j th observed behavior; q is the number of common factors. I used a Scree
plot analysis to determine the number of significant factors (Stevens 1992), and the
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. factors underwent oblique promax rotation. I then estimated the factor loading scores for
the significant factors (two factors for both VT and MT experiments) using the factor
scoring coefficient determined by PROC FACTOR. Scores were calculated for each
individual fly by multiplying the score coefficient by the raw data (PROC SCORE). I fit
a linear model to these scores for each significant factor with family, day, and trial nested
in day as independent variables.
Results
VT Experiment
Male Courtship Progression
Full-sibling families differed in their temporal patterns of courtship during the
behavioral trials. In the repeated-measures analysis of males exposed to virgin females,
the family*time interaction was significant, and the family* time2 effect was marginally
non-significant (Table 2.2). The family effect was not significant, indicating that families
did not have significant genetic differences at the beginning of the trial. Both linear and
quadratic effects of time were significant, indicating a robust quadratic temporal pattern
of behavior variation during the trials. Figure 2.2a illustrates this pattern by showing the
mean MCP scores for eight families (mean MCP scores for all VT families are provided
in appendix B). Day and trial-within-day effects were also significant, demonstrating that
male courtship behavior was influenced by daily environmental effects and within-day
temporal effects.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Comparing MCP to Other Metrics
There were significant family differences for Cl proxy, and for copulation
occurrence, latency, and duration (Table 2.3). The temporal effect of day was significant
for Cl as well as copulation latency and duration; trial-within-day was significant only for
copulation occurrence. The composite measure of Cl was significantly related to
copulation occurrence (rs = 0.51, n=80, p< 0.001), copulation duration (rs = 0.76, n=80,
p< 0.001), and latency to copulation (rs = -0.58, n=80, p< 0.001).
The Scree analysis indicated two important canonical factors for VT families
(eigenvalues >1.5, total variance explained: 58.6%). Factor 1 was dominated by positive
loadings of wing vibration, licking, and attempted copulation and a moderate negative
loading for motionless. Factor 2 was dominated by a large positive loading for
copulation, and a strong negative loading for motionless (Table 2.4). Families differed
significantly for Factor 2, and the effect of day was also significant; however, none of the
effects were significant for Factor 1 (Table 2.3).
MT Experiment
Male Courtship Progression
Results of the MCP analysis for males exposed to mated females were similar to
those for virgin females. Families differed significantly in how behavior changes over
time, but not in the baseline behavior at time 0 (Table 2.2). In this experiment, families
differed in the quadratic and in the linear effects o f time (Figure 2.2b, mean MCP scores
for all MT families are provided in appendix C). As in the VT experiment, time effects
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were highly significant. Day was not significant in this analysis, although trial within day
was marginally significant.
Comparing MCP to Other Metrics
In contrast to the results for the VT experiment, there were no significant
differences between families for any of the univariate metrics (Table 2.3). This is also
very different from the results of the MCP analysis, as that metric successfully detected
significant behavioral differences between families when these other metrics did not.
Day was significant for copulation occurrence, as in the VT experiment. The effect of
trial within day was not significant for any analyses.
The Scree analysis from the factor analysis indicated two important factors
(eigenvalues >0.5, total variance explained: 78.4%). As in the VT experiment, wing
vibration, licking, and attempted copulation loaded strongly positively on Factor 1 (Table
2.4). However, Factor 2 is mainly influenced by positive loadings of following and wing
vibration and negative loadings of motionless and copulation. The difference in Factor 2
loadings from those seen in the VT experiment (where Factor 2 reflected a strong positive
loading of copulation) is likely due to rarity of copulations in the trials with previously
mated females. Also unlike the VT experiment, families did not differ significantly for
either Factor (Table 2.3). The only significant effect in the analysis of MT factors was a
significant effect of day on Factor 1. For a summary of results for MCP compared to
other analyses, refer to Table 2.5.
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion
I found substantial genetic variation in male courtship behavior when males are
exposed either to virgin or to previously mated females. This is a rare demonstration of
natural genetic variation in male courtship behavior towards previously mated females in
Drosophila. Families differed mainly in the rate at which high levels of courtship
intensity are achieved in both mated- and virgin female experiments. The primary
difference between the two experiments is that in mated-female experiments, even the
families with the highest-intensity behavior did not have many individuals advance
beyond “licking” on the MCP scale. This result suggests that female receptivity is
necessary for males to progress to high levels of courtship intensity. Females in the MT
experiments had mated ~ 24 hours previous to their mating trial, and were likely not
receptive to mating (Markow 2002).
In the MT experiment, genetic variation in behavior reported here does not seem
to reflect variation in the occurrence of courtship conditioning (“Courtship conditioning”
refers to the substantial decrease in Drosophila male courtship behavior after extended
exposure to non-receptive females (Siegel and Hall 1979; Gailey et al. 1984; Reif et al.
2002)). Rather, there appears to be more variation in the rate of increase in courtship
intensity early in the mating trial than in the rate of decline in intensity after 20 minutes
(see Figure 2.2b). This could potentially correspond to variation in the ability of different
families to detect the presence of a female, receive and process stimuli, and respond
appropriately in a timely manner.
Although all behavioral trials were confined to two hours in the morning, and
environmental conditions for all males were highly controlled, I detected significant
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. among-day and within-day variation in behavior in both experiments. Among-day
variation suggests that subtle effects such as daily fluctuations in relative humidity in the
laboratory (not controlled in these experiments) can have significant effects on behavior.
Within-day effects suggest that circadian variation in male courtship behavior may be
occurring over quite short time scales. In fact, in both the virgin- and mated-female
target experiments, the average MCP score decreases by approximately 0.5 from trial 1
(approximately 8 AM-8:30 AM) to trial 4 (approximately 9:30 AM-10 AM), indicating a
decrease in courtship behavior in late morning. This demonstrates the importance of
accounting for the effects of environmental and circadian variation when performing
statistical analyses of behavior, even when efforts are made to experimentally control for
them.
My method appears robust for quantifying differences among individuals in the
pattern of courtship behavior in Drosophila. The MCP scale was more successful at
detecting these differences than other metrics, especially in the mated-female experiment
(Table 2.5). For example, although a metric such as “latency to copulation” can provide
insight into temporal patterns of behavior (Moehring and Mackay 2004; Mackay et al.
2005), it is not useful when very few males actually achieve copulation, as in our MT
experiment. This result highlights the importance of quantifying subtle differences in
courtship behavior. The MCP was the only metric able to detect significant variation for
males placed with mated females, which is usually what males would encounter in the
wild, and hence where variation in courtship influencing mating success may have a
significant impact on evolution of behavior.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. One limitation of this new method is that the 30-second interval between
instantaneous scan sampling will miss some behavioral transitions. In Bastock and
Manning’s report of male courtship behavior (1955), orienting behavior had a durational
range of 6-15 seconds, while wing vibration had a range of 2-16 seconds. Markow and
Hanson (1981) reported that chasing behavior had a mean duration of 3 seconds.
However, the MCP method still successfully captures variation in behavioral patterns,
and can do so for a much larger number of flies than any other method that would capture
all of these transitions. This assay is straightforward and scalable to the large sample
sizes that are needed for genetic analysis.
This experiment reveals the importance of examining overall patterns of courtship
behavior as opposed to using summary metrics or focusing on specific behaviors from a
sequential series. Using the appropriate analysis of behavioral patterns allowed me to
detect significant genetic differences in courtship when other methods would have
detected no difference, especially in experiments with mated females as targets. This
method is also relatively easy to use and provides a detailed account of behavior in a
high-throughput assay. Using the MCP scale, I detected the presence of genetic variation
in behavior in a natural population using a sample size large enough to extend this to
behavior-genetic analysis. I can now use this information to move forward in the process
of understanding the genetics of courtship behavior by examining the molecular basis of
this variation, identifying why this variation is being maintained, and looking for
mechanisms which may be under sexual selection.
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements
I would like to thank the University of Illinois Program in Ecology, Evolution, and
Conservation and the Francis M. and Harlie M. Clark research grant for financial support
for this research. I would also like to thank M. M. Reedy and the members of the Hughes
lab for help with the flies, and R. M. Reynolds, J. M. Dmevich, A. L. Toth, A. M. Bell,
R. Fuller, H. M. Robertson, A. V. Suarez, and G. E. Robinson for their useful comments
on this manuscript.
List of References
Andersson MB (1994) Sexual Selection. Princeton, N. J.: Princeton University Press. Aspi J, Hoikkala A (1993) Laboratory and natural heritabilities of male courtship song characters in Drosophila montana and D. Littoralis. Heredity 70(4): 400-406. Balakireva M, Stocker RF, Gendre N, Ferveur JF (1998) Voila, a new Drosophila courtship variant that affects the nervous system: Behavioral, neural, and genetic characterization. Journal o f Neuroscience 18(11): 4335-4343. Bastock M, Manning A (1955) The courtship of Drosophila melanogaster. Behaviour 8: 7-110. Beaver LM, Giebultowicz JM (2004) Regulation of copulation duration by period and timeless in Drosophila melanogaster. Current Biology 14(16): 1492-1497. Boake CRB, Konigsberg L (1998) Inheritance of male courtship behavior, aggressive success, and body size in Drosophila silvestris. Evolution 52(5): 1487-1492. Bouletreau J (1978) Ovarian activity and reproductive potential in a natural population of Drosophila melanogaster. Oecologia (Berl) 35: 319-342. Casares P, Carracedo MC, Miguel ES, Pineiro R, Garciaflorez L (1993) Male mating speed in Drosophila-melanogaster - differences in genetic architecture and in relative performance according to female genotype. Behavior Genetics 23(4): 349-358.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Drapeau MD, Long AD (2000) The Copulatron, a multi-chamber apparatus for observing Drosophila courtship behaviors. Drosophila Information Services 83: 194-196. Etges WJ (2002) Divergence in mate choice systems: Does evolution play by rules? Genetica 116(2-3): 151-166. Ferveur JF, Jallon JM (1993) nerd, a locus on chromosome-III, affects male reproductive-behavior in Drosophila-melanogaster. Naturwissenschaften 80(10): 474-475. Gailey DA, Jackson FR, Siegel RW (1984) Conditioning mutations in Drosophila- melanogaster affect an experience-dependent behavioral-modification in courting males. Genetics 106(4): 613-623. Gefen E, Marlon AJ, Gibbs AG (2006) Selection for desiccation resistance in adult Drosophila melanogaster affects larval development and metabolite accumulation. The Journal o f Experimental Biology 209: TBA. Gleason JM, Ritchie MG (2004) Do quantitative trait loci (QTL) for a courtship song difference between Drosophila simulans and D. sechellia coincide with candidate
genes and intraspecific QTL? Genetics 166(3): 1303-1311. Greenspan RJ, Ferveur JF (2000) Courtship in Drosophila. Annu Rev Genet 34: 205-232. Gromko MH (1987) Genetic constraint on the evolution of courtship behavior in Drosophila-melanogaster. Heredity 58: 435-441. Gromko MFI, Markow TA (1993) Courtship and remating in field populations of Drosophila. Animal Behavior 45(2): 253-262. Gromko MFI, Briot A, Jensen SC, Fukui HH (1991) Selection on copulation duration in Drosophila-melanogaster - predictability of direct response versus unpredictability of correlated response. Evolution 45(1): 69-81. Hall JC (1994) The mating of a fly. Science 264(5166): 1702-1714. Hoffmann AA (1999) Is the heritability for courtship and mating speed in Drosophila (fruit fly) low? Heredity 82: 151-157. Kulathinal RJ, Singh RS (2004) The nature of genetic variation in sex and reproduction- related genes among sibling species of the Drosophila melanogaster complex. Genetica 120(1-3): 245-252.
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Littell RC, Stroup WW, Freund RJ (2002) SAS for Linear Models, Fourth Edition. Cary, N. C.: SAS Institute Inc. Mackay TFC, Heinsohn SL, Lyman RF, Moehring AJ, Morgan TJ et al. (2005) Genetics and genomics of Drosophila mating behavior. Proceedings o f the National Academy o f Sciences o f the United States o f America 102: 6622-6629. Markow TA (2002) Perspective: Female remating, operational sex ratio, and the arena of sexual selection in Drosophila species. Evolution 56(9): 1725-1734. Markow TA, Hanson SJ (1981) Multivariate analysis of Drosophila courtship. Proceedings o f the National Academy o f Sciences o f the United States o f America 78(1): 430-434. Markow TA, Castrezana S, Pfeiler E (2002) Flies across the water: Genetic
differentiation and reproductive isolation in allopatric desert Drosophila. Evolution 56(3): 546-552. Martin P, Bateson P (1993) Measuring Behaviour: An introductory guide. Cambridge: Cambridge University Press. Moehring A, Mackay T (2004) The quantitative genetic basis of male mating behavior in Drosophila melanogaster. Genetics 167(3): 1249-1263. Partridge L, Hoffmann A, Jones JS (1987) Male size and mating success in Drosophila- melanogaster and Drosophila-pseudoobscura under field conditions. Animal Behaviour 35: 468-476. Reif M, Linsenmair KE, Heisenberg M (2002) Evolutionary significance of courtship conditioning in Drosophila melanogaster. Animal Behaviour 63(Part 1): 143-155. Ritchie MG, Kyriacou CP (1994) Genetic-variability of courtship song in a population of Drosophila-melanogaster. Animal Behaviour 48(2): 425-434. Ritchie MG, Gleason JM (1995) Rapid evolution of courtship song pattern in Drosophila-willistoni sibling species. J Evolution Biol 8(4): 463-479. Ryner LC, Goodwin SF, Castrillon DH, Anand A, Villella A et al. (1996) Control of male sexual behavior and sexual orientation in Drosophila by thefruitless gene. Cell 87(6): 1079-1089. SAS Institute (2006) SAS online documentation v. 9.1. Available: http://support.sas.com/91doc/docMainpage.isp. Accessed 2006.
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sawamura K, Tomaru M (2002) Biology of reproductive isolation in Drosophila: Toward a better understanding of speciation. Population Ecology 44(3): 209-219. Service P, Rose M (1985) Genetic covariation among life-history components- the effect of novel environments. Evolution 39(4): 943-945. Siegel RW, Hall JC (1979) Conditioned responses in courtship behavior of normal and mutant Drosophila. Proceedings o f the National Academy o f Sciences o f the United States o f America 76(7): 3430-3434. Stamenkovicradak M, Partridge L, Andjelkovic M (1992) A genetic correlation between the sexes for mating speed in Drosophila-melanogaster. Animal Behaviour 43(3): 389-396. Stevens J (1992) Applied Multivariate Statistics for the Social Sciences. Hillsdale, N.J.: L. Erlbaum Associates. Stirling DG, Reale D, Roff DA (2002) Selection, structure and the heritability of behaviour. J Evolution Biol 15: 277-289. Yamamoto D, Nakano Y (1998) Genes for sexual behavior. Biochemical & Biophysical Research Communications 246(1): 1-6. Zar JH (1996) Biostatistical Analysis. Upper Saddle River, N.J.: Prentice Hall.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLES
Table 2.1. Male Courtship Progression Scale
Score Behavior
1 Motionless
2 M oving
3 Oriented toward female
4 Following/chasing female
5 Wing vibration
6 Licking genitalia
7 Attempted copulation
8 Copulation
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.015 67.58 (48) 0.174 0.033 16.39(12) 35.3 (1) Effect 17.39(1) 10.45(1) 49.28 (14) 71.61 (48) <0.0001 <0.0001 Family Family*Time Family*(Time)2 Time Time*Time Day Trial(Day) 0.633 0.012 0.042 67.72 (77) 105.12 (79) 97.74 (79) 70.26 (75) 108.9 (78) 100.9 (78) 84.27 (1) 2.2. Generalized linear model analysis ofMCP VT ExperimentVT P-valueMTExperiment 0.766P-value 0.026 0.075 <0.0001 0.001 <0.0001 X2 (df) X(df) Table
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.3. Statistical tests for univariate and multivariate models
E ffect1
Family Day Trial(Day)2
Trait
VT Experiment
Cl 2.0 (77,451) *** 1.81 (14,451)* 0.94 (48,451)
Copulation Occurrence2 105.8 (79) * 17.4(14) 86.5 (48) **
Copulation Duration 1.5 (77,374) * 1.89(14,374)* 1.31 (48,374)
Latency to Copulation 1.4 (77,450) * 2.24(14,450) ** 1.22 (48,450)
Factor 1 1.0 (77,451) 0.74(14,451) 0.97 (48,451)
Factor 2 1.8 (77,451)** 2.22 (14,451)* 1.28 (48,451)
MT Experiment
Cl 0.98 (75,388) 1.46(12,388) 0.92 (48,388)
Copulation Occurrence 88.3 (78) 34.3 (12) *** 50.3 (48)
Copulation Duration 1.1 (2,5) 0.20 (3,5) N/A3
Factor 1 0.94 (75,388) 1.85 (12,388)* 1.02 (48,388)
Factor 2 1.2 (75,388) 0.89 (12,388) 0.98 (48,388)
* p<0.05; ** p<0.01; *** p<0.001
1 - Family, day, and trial(day) effects tested using Type III F-statistics
2 - Copulation occurrence tested usingy2
3 - Model possessed infinite likelihood with trial(day) effect;
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.4. Multifactorial factor pattern analysis
VT Experiment
Behavior Factor 1 Factor 2
Motionless -0.4806 -0.7057
Moving 0.0149 -0.3832
Orienting 0.3151 -0.4454
Following 0.405 0.0437
Wing Vibration 0.8235 -0.187
Licking 0.8608 -0.2143
Attempted Copulation 0.8271 -0.0622
Copulation -0.2886 0.9413
MT Experiment
Behavior Factor 1 Factor 2
Motionless -0.6645 -0.4949
Moving -0.6604 0.3268
Orienting -0.0009 0.1167
Following -0.0563 0.6823
Wing Vibration 0.648 0.5704
Licking 0.7381 0.1466
Attempted Copulation 0.6117 -0.2621
Copulation 0.0063 -0.5013
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.5. Summary of success of MCP and other methods for detecting presence of
genetic variation
Method VT Experiment MT Experiment
MCP * *
Cl Proxy *** N.S.
Copulation Occurrence * N.S.
Copulation Duration * N/A
Latency to Copulation * N/A
Factor Pattern 1/ Factor Pattern 2 N.S./** N.S./N.S.
* p<0.05; ** p<0.01; *** pO.OOl
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIGURES
Figure 2.1. Courtship observation device modeled after Drapeau and Long (2000). Top
row of device shows male and female chambers separated by thin opaque sheet. Bottom
row of device shows full mating arenas comprised of both male and female chambers.
Smaller side chambers are filled with media to prevent dessication during trials.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.2. As an example of male courtship progression, average male courtship progression score is plotted over course of trial for four “High MCP” and four “Low MCP” families (a) placed with virgin female targets and (b) placed with mated female targets. A total of eighty families were analyzed for each experiment. Only the first fifteen minutes of each trial were used in generalized linear model analysis of MCP (see text).
a.
o L _ _------0.5 3.5 6,5 9.5 12,5 15,5 18,5 21,5 24,5 27.5 30
Time (minutes)
b.
Family
Time (minutes)
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 3: Alteration in expression of courtship genes due to
age and experience in maleDrosophila melanogaster
Abstract
Differences in gene expression can be one target of selection. Genes can exhibit static
expression, where baseline expression levels do not change over the course of adult life.
They may also exhibit dynamic expression, where expression levels change in response
to a non-genotypic factor; for example, age or experience. Here I examine whether genes
previously identified through mutation screening as being associated with male courtship
behavior in the fruit fly,Drosophila melanogaster, have static or dynamic expression in
the heads of genetically identical flies placed in different age/experience scenarios. This
information can give us an understanding of the complexities of expression for genes
potentially involved in variation of male behavior. For this genotype, I found that 16
genes were statically expressed, and 9 genes were dynamically expressed with changes in
response to age. No genes exhibited dynamic expression changes due to experience or
age*experience interactions.
Introduction
Evolution via alterations in gene expression and mutations in regulatory regions
has been proposed as a major contributor to phenotypic variation between species (King
and Wilson 1975). Differences in gene expression both between and within species show
evidence of positive and stabilizing selection (Ferea et al. 1999; Oleksiak et al. 2002;
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rifkin et al. 2003; but see Khaitovich et al. 2004). Substantial variation in expression has
been identified within populations (Oleksiak et al. 2002; Wayne et al. 2004; Hughes et al.
2006); therefore, changes in gene expression could be an underlying cause of phenotypic
variation within populations. One such phenotype is behavior, which can be critical for
individual fitness.
Alteration of the expression of a gene can be one target of selection (Harris-
Warrick 2000; Insel and Young 2000); and in fact, several cases of variation in behavior
can be attributed to differential gene expression. Both baseline differences in statically
expressed genes between individuals and changes in dynamically expressed genes can be
important for the evolution of behavior. One of the most well characterized examples is
theforaging gene. In the fruit fly, Drosophila melanogaster, theforaging gene has two
alleles, f o f and fo rs, which correspond to the rover and sitter foraging phenotypes of
both larvae and adults. The rover allele results in consistently higher mRNA levels of the
fo r gene and higher levels of PKG activity than the sitter allele, in both the juvenile and
adult phases. These differences in mRNA abundance and PKG activity are static, but are
under active selective pressure depending on the ecological conditions that the population
is experiencing (Fitzpatrick et al. 2007).
Like statically expressed genes, dynamically expressed genes can also affect
behavior with important fitness consequences. The ortholog of the foraging gene in the
honeybee, Apis mellifera, is again a well-characterized example. In A. mellifera,
individuals go through a nursing stage during which they care for the hive and brood.
Typically, at 2-3 weeks of age, an individual will transition from nursing to foraging
tasks. Expression of the ortholog to the D. melanogaster gene foraging, Amfor, is
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. significantly higher in foraging individuals than in nurses. In addition, increased PKG
activity causes accelerated transition from nurse to forager (Ben-Shahar et al. 2002).
These dynamic expression changes are integral to the social roles of individuals, and thus
are crucial for the fitness of the hive.
Another example of selection on a dynamically expressed gene is in the cichlid,
Haplochromis burtoni. In this species, the expression of a gene for gonadotropin-
releasing hormone ( GnRH l ) is socially regulated. Dynamic changes in the expression of
GnRHl influence the switch from aggressive to submissive behaviors associated with
social status; these changes in expression can affect the reproductive success of the fish
(White et al. 2002).
Because selection can act on both statically and dynamically expressed genes, a
complete understanding of how gene expression can cause variation in behavior requires
knowledge of how genes are regulated by changes in environment or physiology due to
age or experience. To examine the effects of age and experience on changes in gene
expression, I used male courtship behavior in Drosophila melanogaster , which is one of
the best characterized behavioral patterns in a model organism. Male courtship in the
fruit fly consists of a suite of complex behaviors that are genetically modulated (Hall
1994b). Mutation screening has revealed many genes that are necessary for normal
courtship. However, the alleles produced by this process usually result in null mutations
that have many pleiotropic effects in addition to the disruption of normal courtship
behavior, and alleles with such severe mutations are unlikely to persist at an appreciable
frequency in natural populations (Hall 1994a; Greenspan 1997a).
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Genes that have been identified by mutant screening are known to influence male
courtship behavior, but their temporal expression patterns are often unknown.
Understanding the expression pattern of these genes (further referred to as courtship
deficient, or CD genes) in response to age and environment is critical to investigating
their contributions to the natural variation in male courtship behavior. Here, I was
interested in changes in gene expression of CD genes in the head of adult male flies due
to age or experience. In this situation, if the expression of a CD gene changed either due
to age (temporal regulation) or experience (environmental regulation), I considered it
dynamically expressed. I was particularly interested in up- or down-regulation of CD
genes in reproductively mature males that had experienced courtship; changes in gene
expression only in this age/experience category would indicate that the CD gene
specifically responds to courtship. If a CD gene did not change expression due to age or
experience, I considered it statically expressed. Static expression could indicate that the
CD gene is involved in the development and/or maintenance of the organism. However,
CD genes that appear to be statically expressed in this experiment might have spatially
dynamic expression (for instance, tissue-specific expression), dynamic expression in
response to repeated exposure to females, dynamic expression resulting from copulation,
etc. Thus, I cannot extend my claim of static expression to scenarios outside these
particular age/experience combinations.
I conducted an experiment to investigate age- and experience-dependent
regulation of gene expression involved with courtship. Age and experience are key
factors in male courtship success (Pitnick et al. 1995; Reif et al. 2002), but it is not
known how expression of genes originally identified by mutation screening changes with
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. either variable. There is evidence for alterations of neural physiology due to age or
experience (Strambi et al. 1999; Devaud et al. 2003), and for genetic variation in age-
related male reproductive physiological maturity (Pitnick et al. 1995; Promislow and
Bugbee 2000). Therefore, age and experience do have profound effects on the expression
of genes related to these phenotypes (which may also feasibly be related to courtship
behavior). For CD genes, it would be not be unreasonable to expect expression
differences related to age, experience, or a combination of those factors. I found that age,
but not experience, altered the expression of 9 CD genes, and that the other 16 genes
examined were statically expressed.
Methods
Flies
Flies used in this study were derived from the New Jersey (NJ) stock of D.
melanogaster. This stock originated from 8,000 offspring of 4,000 wild-caught females
collected in Terhune, New Jersey, in 1999 by Valerie Pierce. The stock was maintained
in a large, randomly mating outbred population in the laboratory of Dr. Allan Gibbs. We
obtained 500 of these flies in 2003, where they have been maintained at a population size
of approximately 12,000 individuals with overlapping generations.
To create isogenic lines, virgin females were randomly chosen from the NJ base
stock, and the X, II, and III chromosomes from these flies were extracted by Dr. Jenny
Dmevich using balancer chromosomes and standard Drosophila crossing schemes
(Greenspan 1997b). The balancers used in creation of these lines were FM7a and
T(2;3)A1-W, In(2L)Cy, In(2R)Cy, Cy1 L1: TM2/T(2;3)UbxB18, In(2LR)bwvl, bwvl Sb'.
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Using microsatellite analysis, we confirmed that the lines used in this study were
homozygous at 7 highly variable microsatellite loci spread out across the genome.
Two of the isogenic NJ lines (17 and 77) were crossed to produce genetically
identical, non-inbred males for this experiment. Experimental flies were reared under
controlled density to minimize maternal and common-environment effects. Virgin
females from line 17 were crossed to males from line 77 in plastic bottles containing 25
flies of each sex. First instar larvae were collected from each of 30 different bottles on
agar plates supplied with yeast paste. Two groups of 25 larvae were collected per bottle,
and each group was transferred to a rearing vial supplied with standard commeal media.
The line 17 x line 77 cross produced progeny that were genetically identical, but not
inbred, for the majority of the genome.
Age and experience categories
My aim was to investigate gene expression differences due to age and experience.
Hence, I examined gene expression differences in flies from four categories, representing
two different ages and two different levels of experience. Age/experience categories
included: 3-day old, reproductively mature males (Pitnick et al. 1995) with courtship
experience (mature and experienced, “ME”) or without courtship experience (mature but
naive, “MN”); and 1-1.5 hour old, immature males who had been exposed to mature
females (immature having encountered a female, “IE”) or who had no experience with
females (immature and nai've, “IN”). In the NJ population, reproductively immature
males can begin to exhibit courtship behavior as early as two hours post-eclosion
(E.A.R., personal observation); however, their degree of progression through the
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. courtship behaviors or their subsequent success with females at this age has not been
determined.
To control for non-experimental environmental or circadian differences, tissue
from flies for all categories was collected simultaneously, within a single half-hour
period on a single day. This collection was made at 9:30 AM (2 hours after the light
cycle begins for our stocks), when genes exhibiting circadian expression are relatively
stable (Leloup and Goldbeter 2000) and flies are active (Rosato et al. 1997). Males for
the ME and MN categories were collected as virgins 0-1 hours post-eclosion from the
constant larval density vials on Day 1 using light CO 2 anesthesia, and were housed
individually so that they had no contact with other flies. On Day 2, ME males were
lightly anesthetized with CO2 and placed into the bottom chambers of a mating device.
Previously-mated females from an outbred stock of flies with ebony body coloration were
placed into the top chambers, and the male and female chambers were separated by an
opaque plastic sheet. Previously-mated females were chosen because they are less
receptive to male courtship (Markow 1996), and thus males should be able to experience
courtship, but not copulation. Four devices with ten male chambers each were set up in
this manner, and were placed in incubators overnight so the flies could recover from
anesthesia (N=40). MN males were kept isolated (N=44).
On Day 3 ,1 collected newly emerged virgin males (0-1 hours post eclosion) for
the immature categories using light CO 2 anesthesia. These males were placed
immediately into individual vials (IN males, N=44) or into vials containing one
previously-mated female from the ebony stock (IE males, N=44). As soon as this was
done, the opaque plastic sheet separating the ME males from their females was removed.
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The ME males and IE males were allowed to interact with a female for 30 minutes. 30
minutes was chosen because variation in male courtship progression can already be
detected after a 30-minute observational period (See Chapter 2). If any ME males were
observed copulating, they were not included in further analysis, since I did not want gene
expression differences due to courtship experience to become confounded with
differences due to copulation. After this 30-minute period, all males from all categories
were frozen using dry ice.
ME males endured one additional exposure to light C 02 anesthesia than the other
categories. A significant age*experience effect for a CD gene could thus reflect
expression changes due to C 02 exposure. However, no effect of age*experience was
identified, and thus I do not think this was a confounding factor.
Gene expression
I identified 25 CD genes using both a literature search and a query of FlyBase
(Drysdale et al. 2005). From FlyBase, I extracted 65 genes matching the Gene Ontology
(www.geneontology.org) category of “male courtship behavior” and daughter categories
such as “courtship song”. Using the extensive literature (including Hall 1994b;
Fitzpatrick 2004; Moehring and Mackay 2004), I narrowed this list to 25 genes for which
expression in adult male heads has been documented. I limited this experiment to
expression in the head because I wanted to narrow my focus on genes involved in
nervous system function. The CD genes, their molecular functions, their role in male
courtship behavior, and key citations are listed in Table 3.1. If genes showed no
significant change in expression between any of the age/experience categories, I
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. considered them static. If genes displayed significant changes in expression between age
groups, between males housed individually or with females, or with any age/experience
interaction, I considered them dynamic.
I used quantitative real-time PCR to measure abundance of mRNA for each CD
gene in males from each age-experience category. Forward and reverse primer sets were
designed for each of the 25 CD genes using Primer Express software v2.0 (Applied
Biosystems) with default “Taqman Primer and Probe Sets” settings, or using published
qPCR primers (transformer-2 and doublesex Tarone et al. 2005) (Table 3.2). If a gene
had known male-specific exons or transcripts, primers were designed to specifically
amplify those. Primers were checked for specificity using NCBI BLAST searches.
Heads were dissected from the frozen flies in each age/experience category.
Heads from four individual flies were pooled to yield sufficient total RNA collected for
analysis The number of independent RNA pools assayed (biological replicates) was:
MN=11; ME=9; IN= 11; IE=11. Total RNA was extracted using PicoPure RNA
extraction kits (Arcturus), using the manufacturer’s protocol including treatment with
DNasel (Qiagen). RNA was quantified and checked for purity using a spectrophotometer
(Nanodrop) at 260nm. All samples had a 260/280 nm ratio of 1.8 or greater. 200 ng
RNA was reverse transcribed using a mixture of 2pL lOx first strand Arrayscript buffer
(Ambion), lpL lOmM dNTP mix (Applied Biosystems), 0.2pL RNase inhibitor (Applied
Biosystems), and 0.2pL 200U/pL Arrayscript (Ambion) in DEPC-treated water. As an
exogenous control, I spiked lOng of plant Root Cap Protein I cRNA (RCP1, obtained
from Dr. Thomas Newman) into each reaction. Reactions were incubated at 42°C for 60
minutes, then at 95°C for 5 minutes. The cDNA from one RT reaction could be used to
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. complete qRT-PCR reactions for 9-10 genes. Therefore, three separate RT reactions
were performed for each biological replicate in order to yield enough cDNA to complete
qRT-PCR reactions for all 25 genes.
For each gene, a mixture of 5 pL 2x SYBR Green Master Mix (Applied
Biosystems), 1 pL lOpM F primer, and lpLlO pM R primer was added to 3 pL of 6x
diluted cDNA from each sample. This reaction was performed in triplicate for each
sample (technical replicates). Specific transcripts from each sample were quantified
using the ABI Prism HT7900 sequence detection system (Applied Biosystems). Each
reaction was performed with the default PCR cycle settings for 40 cycles, and a
dissociation curve was added to the final cycle to confirm the absence of primer-dimers
for each gene. A five-fold log-scale dilution standard curve was generated for each gene
using D. melanogaster genomic DNA in order to determine absolute quantification for
each sample. A qRT-PCR reaction for the exogenous RCP1 control gene was performed
for each of the three separate RT reactions used to generate cDNA. This allowed for
standardization of the absolute quantity of expression of the CD genes by accounting for
differences between samples in the RT reactions. A five-fold log-scale dilution standard
curve was generated for RCP1 using RPC1 cRNA.
Statistical Analyses
To standardize the absolute quantities of cDNA for each gene (determined from
the standard curve, in nanograms) I used the exogenous RCP1 cRNA control to account
for differences in efficiency of reverse transcription between samples (for details about
RCP1 and molecular methods, see above). This standardization was accomplished by
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dividing the raw absolute quantity of a gene (in nanograms) for each sample by the raw
absolute quantity of RCP1 (in nanograms) for that sample. After standardization with the
exogenous control, all expression data were log-transformed to normalize the residuals in
order to meet the necessary assumptions for the statistical tests.
To evaluate the overall effects of age and experience, I performed a multivariate
analysis of variance (MANOVA) on expression of all genes simultaneously using R
statistical software v. 2.5 (R Development Core Team, 2007). I tested for the effects of
age, experience, and their interaction as in the ANOVA for individual genes. I used the
Pillai trace statistic to approximate the Type-III F-statistic in my MANOVA, which uses
eigenvalues to compare the error sums of squares to the sums of squares of the
hypotheses (Johnson 1998).
To examine the ability of expression profiles to appropriately classify flies into
the age/experience categories, I used a linear discriminant analysis (Everitt 2005). The
linear discriminant analysis is equally as effective as alternative parametric and non-
parametric versions of classification analyses (Meyer et al. 2002), especially in cases
where only a few canonical factors explain most of the variation (Johnson 1998). In
addition, it can allow the researcher to interpret variables and visualize distances between
populations on a plane (Johnson 1998), which many other classification methods cannot
do. To determine the number of significant linear discriminant factors in my expression
data, I used SAS PROC CANDISC (SAS Institute, 2007). The eigenvalues showed that
LD1 was highly significant (F75_28= 5.62; P<0.0001), while LD2 was marginally non
significant (F48> 2o= 1.88; P= 0.062) and LD3 was not significant (F23,u=1.37; p=0.3).
Thus, any discussion of LD3 should be interpreted with caution.
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I next performed linear discriminant analyses using the “Ida” function in R
(MASS package). I then used the “predict” function to classify individual flies based on
resubstitution using the linear discriminate scores for all three discriminant factors, and
cross-validated the classification using the “CV” command in the “Ida” function. CV
performs a leave-one-out cross-validation of the ability of the LD loading scores to
accurately predict the class of each individual, and is a more conservative estimate than
resubstitution. Finally, I visualized the expression patterns of the age/experience
categories using single-factor hierarchical clustering on normalized data with Genesis
software (Stum et al. 2002).
To determine the significance of effects of age, experience, and the interaction
between these two factors on the expression of individual genes, I performed a fixed-
effect ANOVA (SAS PROC MIXED, SAS Institute, 2007) on the normalized data. Each
gene was fit to a model: y = p + a + x + a*x + e, where y was the log-transformed
standardized quantity of cDNA for that a particular sample, a was the fixed effect of age,
x was the fixed effect of experience, a*x was the interaction effect between age and
experience, and e was the random error (Type III SS). False discovery rate (FDR) was
calculated using the method described in Benjamini and Hochberg (1995) for age,
experience, and age*experience to control for multiple hypothesis tests.
Results
The MANOVA showed a highly significant effect of age, a non-significant effect
of experience, and a marginally non-significant interaction of age*experience on gene
expression differences between the age/experience categories (Table 3.3). The linear
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. discriminant analysis yielded three linear discriminate factors. LD1 accounted for 93.3%
of the total variance, while LD2 and LD3 accounted for 4.8% and 1.9%, respectively (for
linear discriminant loadings see Table 3.4).
Plotting the linear discriminant scores for individuals in each age/experience
category revealed that the majority of the variation in LD1 is explained by differences in
gene expression related to age (Figure 3.1 and 3.2). However, LD1 also appears to be
using variation due to both age and age*experience to appropriately classify flies into
categories. Genes with positive LD1 loading scores represent either higher expression in
mature males, or a more complex interaction of age*experience. Genes with negative
LD1 loading scores tend to represent higher expression in immature males; in fact, the
genes with the largest negative LD1 scores all have higher expression in immature
categories. LD2 appears to discriminate between MN/IE males and ME/IN males,
indicating that certain genes are expressed differently depending on interactions between
age and experience (Figure 3.1). Finally, LD3 discriminates between expression patterns
of males who have and have not encountered females (Figure 3.2). LD2 and LD3 should
be interpreted cautiously, as they account for very little of the overall variance in gene
expression. Predicting classes of individuals using resubstitution of the linear
discriminate scores had an extremely high success rate (ME 100%; MN 100%; IE 100%;
IN 88.9%). The more conservative cross-validation of the linear discriminate
classification yielded similar results for the mature age groups (ME 71.4%; MN 70%),
but was less reliable for the immature age groups (IE 54.5%; IN 44.4%). The combined
resubstitution and cross-validation results confirm that the linear discriminant loadings
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. are appropriately classifying individuals based on their genes expression profiles in the
majority of instances.
The single factor hierarchical clustering showed a tree separating experimental
groups first by age, then by experience (Figure 3.3). There were three interesting gene
clusters. Genes that had higher expression in immature vs. mature males had a distinct
cluster, while genes that had higher expression in mature vs. immature males separated
into two other clusters. One cluster grouped genes with differences mostly due to age,
much like the immature v. mature male cluster. The other cluster exhibited subtle
expression differences between categories that were not associated with just age or
experience, perhaps picking up on the marginally non-significant effect of the
age*experience interaction.
Analysis of variance on individual genes revealed that age had a significant effect
on expression of nine of the genes, but that there were no significant effects of experience
or age*experience interaction before correction for multiple tests (Table 3.5; raw means
and standard deviations for each CD gene are displayed in Figure 3.4). The genes
exhibiting static expression do not show appear to show communality, i.e. they do not
have uniformly lower (or higher) absolute expression levels than the significant genes
(log-transformed expression levels for all genes is reported in appendix D). Therefore,
the lack of effect of age, experience, or a combination of both is probably not due to a
lack of statistical power or unreliable quantification of genes with low (or high)
expression.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion
I found that courtship deficient genes are both statically and dynamically
expressed. Nine CD genes contributing to normal male courtship behavior had altered
temporal expression during the first three days of adult life. This was apparent whether I
analyzed genes individually or as a group. Whether or not a male had encountered (or
even courted) a female did not significantly alter gene expression, although the
MANOVA and linear discriminant analysis demonstrated a slight trend indicating that
the interaction of age and experience might alter the expression of several of the genes.
My results indicate that these genes do not exhibit dynamic expression due to the
immediate effects of experience with a female. However, it is possible that some of these
genes may exhibit altered expression several hours after encountering a female, or after
repeated exposure to a female.
The genes that are contributing most heavily to the age-related variation in
expression are: dsx, eg, fru, K r-hl, pie, pros, tko, to, andy. These genes exhibit dynamic
temporal expression, but do not all change in the same direction or magnitude. Three of
these genes are down-regulated in later ages (eg, pie, y); this could indicate that they are
important in early adult development, or that their down-regulation is necessary for
normal adult function, y and pie are involved in pigmentation (reviewed in Wittkopp et
al. 2003), which perhaps accounts for their heightened expression in immature males.
Those genes that are up-regulated in mature males (dsx, fru, Kr-hl, pros, tko, to)
may require increased expression for normal courtship behavior. Many of these genes
are transcription factors, indicating that there may be genes downstream of these CD
genes that are also involved in male courtship. Three of these genes, dsx, fru . and to, are
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. part of the sex determination cascade, most of which is completed by adulthood
(reviewed in Greenspan and Ferveur 2000; also see Dauwalder et al. 2002). However,
for each of these genes, expression has been found in heads of wild-type adult males (fru:
Ryner et al. 1996; to: Dauwalder et al. 2002; dsx: Lee et al. 2002). Presence of
expression in adults, which I have now determined to by dynamically altered by age,
supports the idea that fru, dsx, and to exhibit as yet unknown pleitropic effects during
adulthood.
The sixteen remaining genes were statically expressed across age and experience
level. It is possible that subtle differences in expression of these genes within individuals
does not affect courtship behavior, but rather it is the absence of a normal protein product
during development and/or in adulthood that alters male courtship (hence the
identification of the gene using mutation screening procedures). The static expression of
these genes could indicate that they are functionally constrained. Although these genes
are statically expressed, differences in baseline levels of expression in individual flies
may have a profound effect on courtship behavior. Additional testing on individuals with
different genetic backgrounds is necessary to examine the possibility that different
genotypes have different baseline expression for these static genes. Additionally, as
discussed before, these genes may exhibit dynamic expression under different
circumstances or in different tissues.
Courtship deficient genes involved in male courtship behavior in D. melanogaster
appear to be mainly statically expressed. This could perhaps be related to constraints in
expression of genes involved with nervous system function (i.e. Galis 1999; Smith 2006);
however, genes expressed in the brain that exhibit altered expression associated with
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. behavioral modification are less constrained than originally thought (i.e. Whitfield et al.
2003; Velarde et al. 2006). The genes that were dynamically expressed changed with age
and perhaps corresponded to the physiological, reproductive maturity of the males; it
would be interesting to see how reproductive maturity alters male courtship behavior.
Finally, there does not appear to be immediate dynamic changes in expression related to
exposure to stimuli from a female or from courtship itself. Thus, for these CD genes,
experience with a female does not have an immediate effect on expression. It is possible
that genotypes other than this one could have stronger age*experience changes in
expression; if so, and if this contributes to phenotypic differences, it could be a target of
selection related to phenotypic plasticity.
Acknowledgements
I would like to thank the National Science Foundation (NSF DEB 0296177), University
of Illinois Program in Ecology and Evolutionary Biology, the Animal Biology Banks
Award, and the Francis M. and Harlie M. Clark Research Grant for funding for this
project. Thanks to G. E. Robinson for the use of his laboratory and equipment. I am
indebted to F. E. Miguez and A. L. Toth for their description of linear discriminant
analysis and R code. I would especially like to thank M. M. Reedy for her help with the
flies, T. Newman for his help with qRT-PCR troubleshooting, and G. Davis for help with
primer design. Thanks to G. E. Robinson, H. M. Robertson, A. V. Suarez, A. L. Toth and
R. M. Reynolds for their helpful commentson this manuscript.
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of References
Anholt RRH, Mackay TFC (2004) Quantitative genetic analyses of complex behaviours in Drosophila. Nat Rev Genet 5(11): 838-849. Baker BS, Taylor BJ, Hall JC (2001) Are complex behaviors specified by dedicated regulatory genes? Reasoning from Drosophila. Cell 105(1): 13-24. Ben-Shahar Y, Robichon A, Sokolowski MB, Robinson GE (2002) Influence of gene action across different time scales on behavior. Science 296(5568): 741-744. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal o f the Royal Statistical Society Series B (Methodological) 57(1): 289-300. Billeter JC, Goodwin SF, O'Dell KMC (2002) Genes mediating sex-specific behaviors in Drosophila. Advances in genetics, vol 47. San Diego: Academic Press Inc. pp. 87-116.
Dauwalder B, Tsujimoto S, Moss J, Mattox W (2002) The Drosophila takeout gene is regulated by the somatic sex-determination pathway and affects male courtship behavior. Gene Dev 16(22): 2879-2892. Devaud JM, Acebes A, Ramaswami M, Ferrus A (2003) Structural and functional changes in the olfactory pathway of adult Drosophila take place at a critical age. J Neurobiol 56(1): 13-23. Diagana TT, Thomas U, Prokopenko SN, Xiao B, Worley PF et al. (2002) Mutation of Drosophila homer disrupts control of locomotor activity and behavioral plasticity. Journal o f Neuroscience 22(2): 428-436.
Drapeau MD (2003) A novel hypothesis on the biochemical role of the Drosophila yellow protein. Biochemical and Biophysical Research Communications 311(1): 1-3. Drapeau MD, Cyran SA, Viering MM, Geyer PK, Long AD (2006) A cw-regulatory sequence within theyellow locus o f Drosophila melanogaster required for normal male mating success. Genetics 172(2): 1009-1030. Drysdale R, Crosby M, Consortium TF (2005) Flybase: Genes and gene models. Nucleic Acids Research 33: D390-D395 http://flvbase.org/.
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Emmons SW, Lipton J (2003) Genetic basis of male sexual behavior. J Neurobiol 54(1): 93-110. Everitt B (2005) An R and S-Plus companion to multivariate analysis', Casella G, Fienberg S, Olkin I, editors. London: Springer-Verlag London Limited. Ferea TL, Botstein D, Brown PO, Rosenzweig RF (1999) Systematic changes in gene expression patterns following adaptive evolution in yeast. Proceedings of the National Academy o f Sciences o f the United States o f America 96(17): 9721-9726. Fitzpatrick M (2004) Pleiotropy and the genomic location of sexually selected genes. American Naturalist 163(6): 800-808. Fitzpatrick MJ, Feder E, Rowe L, Sokolowski MB (2007) Maintaining a behaviour polymorphism by frequency-dependent selection on a single gene. Nature 447(7141): 210-U215. Gaines P, Tompkins L, Woodard CT, Carlson JR (2000) quick-to-court, a Drosophila mutant with elevated levels of sexual behavior, is defective in a predicted coiled- coil protein. Genetics 154(4): 1627-1637. Galis F (1999) Why do almost all mammals have seven cervical vertebrae? Developmental constraints, hox genes, and cancer. Journal o f Experimental Zoology 285(1): 19-26. Greenspan RJ (1997a) A kinder, gentler genetic analysis of behavior: Dissection gives way to modulation. Curr Opin Neurobiol 7(6): 805-811. Greenspan RJ (1997b) Fly pushing: The theory and practice o f drosophila genetics. Plainview, N.Y.: Cold Spring Harbor Laboratory Press. Greenspan RJ, Ferveur JF (2000) Courtship in Drosophila. AnnuRev Genet 34: 205-232. Grosjean Y, Savy M, Soichot J, Everaerts C, Cezilly F et al. (2004) Mild mutations in the pan neural gene prospero affect male-specific behaviour in Drosophila melanogaster. Behav Process 65(1): 7-13. Grozinger CM, Sharabash NM, Whitfield CW, Robinson GE (2003) Pheromone- mediated gene expression in the honey bee brain. Proceedings of the National Academy of Sciences of the United States of America 100: 14519-14525.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hall JC (1994a) Pleiotropy o f behavioral genes. In: Greenspan RJ, Kyriacou CP, editors. Flexability and constraint in behavioral systems: John Wiley & Sons, LTD. pp. 15-27. Hall JC (1994b) The mating of a fly. Science 264(5166): 1702-1714. Harris-Warrick RM (2000) Ion channels and receptors: Molecular targets for behavioral evolution. J Comp Physiol A 186(7-8): 605-616. Hughes KA, Ayroles JF, Reedy MM, Dmevich JM, Rowe KC et al. (2006) Segregating variation in the transcriptome: Cis regulation and additivity of effects. Genetics 173(3): 1347-1364. Insel TR, Young LJ (2000) Neuropeptides and the evolution of social behavior. Curr Opin Neurobiol 10(6): 784-789. Johnson DE (1998) Applied multivariate methods for data analysts. Pacific Grove, CA: Brooks/Cole Publishing Company. Khaitovich P, Weiss G, Lachmann M, Hellmann I, Enard W et al. (2004) A neutral model of transcriptome evolution. Plos Biology 2(5): 682-689. King M-C, Wilson AC (1975) Evolution at two levels in humans and chimpanzees. Science 188(4184): 107-116. Lam BJ, Bakshi A, Ekinci FY, Webb J, Graveley BR et al. (2003) Enhancer-dependent 5' -splice site control of fruitless pre-mRNA splicing. Journal o f Biological Chemistry 278(25): 22740-22747. Lee G, Hall JC, Park JH (2002) doublesex gene expression in the central nervous system of Drosophila melanogaster. J Neurogenet 16(4): 229-248. Leloup JC, Goldbeter A (2000) Modeling the molecular regulatory mechanism of circadian rhythms inDrosophila. Bioessays 22(1): 84-93. Markow TA (1996) Evolution o f drosophila mating systems. In: Hecht MKM, R. J.; Clegg, M. T., editor. Evolutionary biology. New York: Plenum Press, pp. 73-106. Meyer D, Leisch F, Homik K (2002)Benchmarking support vector machines. Vienna, AU: Vienna University of Economics and Business Administration. 78. Moehring A, Mackay T (2004) The quantitative genetic basis of male mating behavior in Drosophila melanogaster. Genetics 167(3): 1249-1263.
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Oleksiak MF, Churchill GA, Crawford DL (2002) Variation in gene expression within and among natural populations. Nature Genetics 32(2): 261-266. Orgad S, Rosenfeld G, Greenspan RJ, Segal D (2000) courtless, the Drosophila ubc7
homolog, is involved in male courtship behavior and spermatogenesis. Genetics 155(3): 1267-1280. Pitnick S, Markow TA, Spicer GS (1995) Delayed male maturity is a cost of producing large sperm in Drosophila. Proceedings of the National Academy of Sciences of the United States o f America 92(23): 10614-10618. Promislow DEL, Bugbee M (2000) Direct and correlated responses to selection on age at physiological maturity in Drosophila simulans. J Evolution Biol 13(6): 955-966. Reif M, Linsenmair KE, Heisenberg M (2002) Evolutionary significance of courtship conditioning in Drosophila melanogaster. Animal Behaviour 63(Part 1): 143-155. Rifkin SA, Kim J, White KP (2003) Evolution of gene expression in the Drosophila melanogaster subgroup. Nature Genetics 33(2): 138-144. Rosato E, Piccin A, Kyriacou CP (1997) Circadian rhythms: From behaviour to molecules. Bioessays 19(12): 1075-1082. Ryner LC, Goodwin SF, Castrillon DH, Anand A, Villella A et al. (1996) Control of male sexual behavior and sexual orientation in Drosophila by thefruitless gene. Cell 87(6): 1079-1089. Smith KK (2006) Craniofacial development in marsupial mammals: Developmental origins of evolutionary change. Developmental Dynamics 235(5): 1181-1193. Sokolowski MB (2001) Drosophila: Genetics meets behaviour. Nat Rev Genet 2(11):
879-890. Strambi C, Cayre M, Strambi A (1999)Neural plasticity in the adult insect brain and its hormonal control. International review of cytology - a survey of cell biology, vol 190. San Diego: Academic Press Inc. pp. 137-174. Stum A, Quackenbush J, Trajanoski Z (2002) Genesis: Cluster analysis of microarray data. Bioinformatics 18(1): 207-208. Tarone AM, Nasser YM, Nuzhdin SV (2005) Genetic variation for expression of the sex determination pathway genes in Drosophila melanogater. Genetical Research 86(1): 31-40.
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tauber E, Eberl DF (2001) Song production in auditory mutants of Drosophila'. The role of sensory feedback. J Comp Physiol A 187(5): 341-348. Toivonen JM, O'Dell KMC, Petit N, Irvine SC, Knight GK et al. (2001) technical knockout, a Drosophila model of mitochondrial deafness. Genetics 159(1): 241-254. Velarde RA, Robinson GE, Fahrbach SE (2006) Nuclear receptors of the honey bee: Annotation and expression in the adult brain. Insect Molecular Biology 15(5): 583-595. Villella A, Ferri SL, Krystal JD, Hall JC (2005) Functional analysis of fruitless gene expression by transgenic manipulations of Drosophila courtship. Proceedings of the National Academy o f Sciences o f the United States of America 102(46): 16550-16557. Wayne ML, Pan YJ, Nuzhdin SV, McIntyre LM (2004) Additivity and trara-acting effects on gene expression in male Drosophila simulans. Genetics 168(3): 1413-1420. White SA, Nguyen T, Femald RD (2002) Social regulation of gonadotropin-releasing hormone. J Exp Biol 205(17): 2567-2581. Whitfield CW, Cziko AM, Robinson GE (2003) Gene expression profiles in the brain predict behavior in individual honey bees. Science 302(5643): 296-299. Wittkopp PJ, Carroll SB, Kopp A (2003) Evolution in black and white: Genetic control of pigment patterns in Drosophila. Trends Genet 19(9): 495-504.
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2,3,4 5 3 2,3 Citation(s) Sex determination 2,3,7,9 Courtship conditioning Courtship conditioning Courtship vigor 6 Courtship conditioning G-protein-coupled receptor binding, cyclic-AMP phosphodiesterase activity Courtship conditioning 2,3,4 DNA binding, mRNA binding, neuropeptide hormone activity DNA binding, transcription factor activity Song transmembrane receptor activity Courtship latency 1 mRNA binding, nucleotide bindingubiquitin-protein ligase activity Courtship drive 2,3,4,7,8 transcription factor activity two-component voltage-gatedsensor activity, potassium channel activity ATP binding, calmodulinprotein serine/threonine binding, kinase activity amn dnc ato dsx eag Gene name Gene abbr. Putative Function Aspect Male Courtship 18 wheeler 18w dunce ether ago-go CaMKII* TABLES Table 3.1. Courtship deficient genes amnesiac atonal potatocouch cpo courtless crl doublesex Continued on next page; * Calcium/Calmodulin-dependent protein kinase II
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 7 2,7,14 9 2,3,7 Sex discrimination Sex determination/Song/ Sex discrimination Courtship conditioningSong Song 10,11 Courtship conditioning/ Courtship latency, occurance 1 ligand-dependent nuclear receptorsequence-specific activity, DNA binding, transcription factor activity transcription factor activity receptorbinding calcium ion binding, voltage-gated sodium channel activity transcription factor activitymRNA binding, nucleotide binding, Possibly pheromone detectiontranscription factor/regulator activity 12 Courtship drive 13 iron ionbinding, tyrosine 3-monooxygenase activity protein binding, transcriptionregulator activity unknown Courtship latency/ Sex discrimination Courtship deficient genes Gene abbr. Putative Function Aspect Male Courtship Citation(s) homer eg pros pie Gene name quick-to-court qtc Continued on nextpage; * Kruppel homolog 1; f no on or offtransient A eagle Table 3.1 (Con’t). homer nonAi Kr-hl* paralytic para prospero pale fruitless fru
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18 7,20,21 3,15 Citation(s) Sex determinationSex determination/Song/ 19 Courtship conditioningCourtship conditioningCourtship latency/ Courtship Vigor 2,3,4 2 Male mating success 15- (Toivonen et al. 2001) 18- (Villella et al. 2005) 19- (Lam et al. 2003) 16- (Dauwalder et al. 2002) 17- (Anholt and Mackay 2004) 20- (Drapeau 2003) 21- (Drapeau et al. 2006) 11 - (Fitzpatrick 11 2004) 12- (Grozinger et al. 2003) 10- (Diagana et al. 2002) 13- (Grosjean et al. 2004) 14- (Gaines et al. 2000) 8- (Orgad et al. 2000) 9- (Baker et al. 2001) adenylate cyclase activity, calcium- and receptor binding, structural molecule activity, voltage-gated potassium channel activity structural constituent ofribosome unknownmRNA binding, protein binding Sex determination 16,17 calmodulin-responsive adenylate cyclase activity ribulose-bisphosphate carboxylase activity nucleic acid binding, transcrption factor, mRNA splicing Sex determination/Song protein binding, Courtship deficient genes to tko Sh y 1- 1- (Moehring and Mackay 2004) takeout transformer tra transformer2 tra2 rutabaga rut 2- (Greenspan and Ferveur 2000) 3- (Sokolowski 2001) 4- (Emmons and Lipton2003) 5- (Tauber and Eberl 2001) 6 -(Hall 1994b) 7- (Billeteret al. 2002) Table 3.1 (Con’t). Gene name Gene abbr. Putative Function Aspect Male Courtship technical Shaker knockout yellow 00
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. GG AAATGGGCGACGTGG GTTC TCCATCGGATCAACCAGACA CTCCACACGCTTCTGTTGGAT R-primer GGCTACCCAGCGCTT AAA GT TCT GG AGT CGGT GG AC AAATTCTAGTC GGACCGGTGG GGGCGTTGCCGGATCT CCGACCGATCGCGAAGT AGCGGTGTCCCTGTATCCACT AAGACCGAAAATAATGCCCC AACCCGCTACCC AAGGTCT GATGCTTCACCCGCAAATGGCCGTCT T CGATTG GTTGAT GCGGATGTGCAATGAGAA T TTCGAAGGAGATTTCAGCAATCAAGGCTTAAGCCACCAT G CGT GCGTGCACATCCTCATCCT CGACCAGGAGGT AGTGAGTAC CGAACACCGACAGGGAGAA GAGGAGCCGCTCGTAGGAT CGGTTTT GGGCCAACACTT AACGTGGCTTCCGTTGGT ATGCTGTGGTAGATGTGCTCTGAGT AC TTCGTGAGCACCTTAGCGCATTTCGCGTTTCATACC AGGA ATCGCAGGTACGCTTATG CGCCGCGC GAGA AAATTTATT CT AACTGTTTATTCCGCCAGGT C AT ACTGCCCGGGACTTCCTT AGTTAACGTACTG GGCGCATAC T CG AAC AGGGTAACCG CGCTGGTAT AGCTCGCAACGAACTGAGGTA ATGGCAGTTGAGTGGTAGGT C GTT GG A AGCT C ACT GT CG AT AC AACAT CGACTGT AGCT CA GTTGG GAAGCGTATAT GC CACTTGTTTG 18w CaMKII eag homer nonA Gene F-primer cpo crl Continued onnext page eg dsx amn ato dnc K r-hl pie pros para fru Table 3.2. qRT-PCR primers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. GGAAGTGGTTAACTATTATTGGCT CTCTT AGAGCTTGCC GGTCCTTT AGTC GT ACTTGTGGCGCGCGT AAGA CAACGAGATTT ATTGCGGACTTT ACGCCAGGTCTGACTGATCACAAGT CCATACCATGCTGGCTAACGGTGCGCGGAGTGGGTGT ATGCG ATTCTAC CTT CC AT CT TTGAAGGTGGATCGGATGGTGCGCCATGCGTTAACTACAA GCTTCT CGAATGACT GTGCTGT AAGGTT TGTCGGTG AGAGTT GCAT GAGCCACGGGAATCTATGTGA AGGTAAGCAAAAAGCC AATGGA AAACTTCAGGAGCGATATAGTTA GG TCTGGCGCTAATGGAGC tko tra2 to Geneqtc F-primer R-primer tra rut Sh y Table 3.2 (Con’t). qRT-PCR primers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.3. Multivariate analysis of variance
Effect Pillai(df) Approximate F(df) P-value
Age 0.99(1) 48.28 (25,9) < 0.0001
Experience 0.76(1) 1.17(25,9) 0.43
Age *Experience 0.87(1) 2.46(25,9) 0.08
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.4. Linear discriminant loading scores
Gene LD1 LD2 LD3
18w 1.460 3.443 -4.008 amn -0.499 0.722 1.938 ato 0.405 5.796 -1.522 CaMKlI -3.280 1.468 -6.462 cpo 2.657 -2.307 0.025 crl 3.857 -6.141 -1.590 dnc -2.348 3.310 -3.768 dsx -0.139 3.264 -0.557 eag -0.603 -0.301 5.171
eg -5.366 0.431 4.743 fru -1.706 -3.042 1.617 homer -0.925 -9.294 -0.686 Kr-hl 1.038 -8.492 -5.620 nonA 2.130 -1.062 5.847 para 1.242 -4.383 1.275 pie -5.849 -0.354 -4.589 pros -0.433 0.164 -4.068 qtc -0.813 0.198 1.227 rut 4.429 3.229 8.839
Sh -1.627 2.164 -2.815 tko 2.401 -0.025 -2.184 to 0.279 2.893 -2.159 tra -2.065 5.610 -2.944 tra2 4.493 -1.053 5.162
y -1.458 5.393 -0.309
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.069 0.924 0.433 0.849 Raw P-value 0.523 0.533 0.830 1.63(1,37) 0.210 0.63(1,37) 3.5(1,38) F-V alue 0.42(1,37) 0.4(1,37) 0.01(1,37) 0.05(1,37) 0.02(1,38) 0.900 0.09(1,37) 0.772 0.421 0.04(1,37) 0.219 0.387 0.01(1,38) 0.918 0.684 0.578 0.291 2.05(1,36) 0.161 1.56(1,38) 1.15(1,36) F-Value Raw P-value 0.77(1,38) 0.03(1,37) 0.858 3.25(1,37) 0.080 0.0500.338 0.06(1,37) 0.89(1,37) 0.806 0.353 0.01(1,37) 0.923 0.295 <0001 *** 0.66(1,37) 0.858 0.99(1,37) 0.326 0.0370.2250.009 * 0(1,37) 0.985 0.29(1,37) 0.592 Age Experience Age*Experience 1.13(1,37) 12.36(1,36) 0.001 ** 1.52(1,37) 7.61(1,38) 0.02(1,37) 0.8810.94(1,37) 8.75(1,38) 0.005 * 0.17(1,37) 0.22(1,38) 0.34(1,38) 0.641 0.563 0.03(1,37) 2.49(1,37)4.09(1,37) 0.1230.51(1,37) 0.47947.37(1,37) 0.32(1,37) 0.26(1,37) 0.616 4.7(1,37) 1- Astericks represent significance after FDR correction formultiple hypothesis testing; * p<0.05, ** p<0.01, *** p<0.001 J8w CaMKll eag Gene F-Value Raw P-value1 amn ato cpo crl dnc dsx eg nonA homer Continued on nextpage; Table 3.5. Analysis ofvariance on individual genes Kr-hl fru
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.691 0.746 0.805 0.746 0.540 0.407 0.387 0.753 0.01(1,37) 0.904 0.16(1,37) 0.11(1,37) 0.43(1,37)0.06(1,37) 0.515 0.11(1,37) 0.38(1,37) F-Value Raw P-value 0.77(1,37) Raw P-value 0.854 0.34(1,37) 0.565 0.068 0.384 0.7(1,37) 1.56(1,37) 0.220 1.75(1,37) 0.195 3.54(1,37) 0.03(1,37) 0.78(1,37) 2.29(1,37) 0.139 0.02(1,37) 0.901 <.0001 *** <0001 *** 0.002 ** 0.042 Age Experience Age*Experience 10.65(1,37) 80.57(1,37) 7.38(1,37)306.65(1,37) 0.01 * 0.07(1,37) 0.794 0.11(1,37) 0.743 0.04(1,37) 0.839 0.04(1,37) 0.847 0.02(1,37) 0.891 4.11(1,37) 0.050 2.71(1,37) 0.108 0.09(1,37) 0.772 4.44(1,37) 234.15(1,37) <.0001 *** 0.48(1,37) 0.491 0.1(1,37) 1- Astericks1- represent significance after FDR correction for multiple hypothesis testing; * p<0.05, ** p<0.01, *** p<0.001 tko to tra2 tra rut Gene F-Value Raw P-value1 F-Value Sh qtc Table 3.5 (Con’t). Analysis ofvariance on individual genes pie y pros para
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIGURES
Figure 3.1. Plot of linear discriminant analysis: LD1 v. LD2. “IN”: immature and naive;
“IE”: immature having encountered a female; “MN”: mature and naive; “ME”: mature
with courtship experience.
in 04 O o M&lid u r n o T -10 0 5 10 LD1
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.2. Plot of linear discriminant analysis: LD1 Vs. LD3.
in CO Q ° IN IEje
o T ■10 0 5 10 LD1
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.3. Gene expression profiles of age and experience categories. Single factor hierarchical clustering of expression profiles. Red indicates increased expression, blue indicates decreased expression. Three distinct clusters were found, and are identified by the color bars on the right. Full gene names are listed in Table 3.1.
18w ato tra rvonA dnc Sh tra2 amn crl fru eag cpo rut tko dsx to Krhl pros CAMKII para Homer pie eg V
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ato eag crl IE M N M E 1.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 0.008 0.006 0.004 0.002 ME
cpo dsx = am n IE MN Category IN
Age: p < 0.05
]—
0 4 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.1 . 0.0 0.01 0 051 0.12 0.02 0.01 . 0.08 0.06 0 0.02 0.025 0.015 0 18w dnc CaMKII IE M N M E Expression levels ofindividual genes for each age/experience category. Raw means and standard deviations are IN
0J 0.6 0.5 0.2 0.1 0.0 0.4 1.5 1.0 10 70 2.5 2.0 50 90 0.5 0.0 30 Figure 3.4. reported. Genes with significant differences in expression due to age have the corresponding p-value displayed (see text). 0.0 Xfl s- W> a os O e A e a o x a !*> V u a 4i 3 4 > *p« "© < SO
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MNME - - p a ra Homer IE IN L_J — — 1.2 1.0 1.0 0.8 0.6 0.4 0.2 0.0 4.0 3.0 2.0 0.0 :
fru pros nonA Ili Ili ifc Category Age: p < 0.01 Age: p < 0.01 ---- — 0.1 0.0 0.2 0.3 0.1 0.0 0.12 0.14 0.08 0.06 0.04 0.02 0,75 0.15 0.05 M N ME Age: p < 0.001 Age: p < 0.001 Expression levels ofindividual genes for each age/experience category. eg pie K r-hl Age: p < 0.05 INIE ------—
— —
0 . 0 1 SB 1.0 1.2 0.8 0.6 0.4 0.2 0.01 I 1.0 7.0 5.0 0.0 3.0 0.02 0.01 Figure 3.4 (Con’t). 0.025 0.015 0.005 £ CQ i- DU o e es C s o X/i u a x CA J3 "© X < v© to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. -I tra2 tko IE M N ME IN Age: p < 0.05 0.2 0.1 0.0 0.2 0.1 0.0 0.25 0.15 0.05 0.25 0.15 0.05 Sh Category IN IE MN ME
1.5 1.0 2.5 2.0 0.5 0.0 Age: p < 0.001 Expression levels ofindividual genes for each age/experience category. rut IN IE MN ME Age: p < 0.001 1.4 1.2 0.8 0.4 0.0 Figure 3.4 (Con’t). 0.025 0.005 0.015 E a W> O c S s Oi U a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 4: Contribution of differential gene expression to
natural variation in male Drosophila melanogaster courtship
Abstract
Behaviors important to fitness often exhibit substantial variation, and thus can be
the targets of selection. In order to evaluate what potential these behaviors have to
evolve, it is important to understand what underlying genetic differences are contributing
to this behavioral variation. Drosophila melanogaster can serve as an excellent model to
answer this question; male courtship behavior is complex and well documented, and
many genes have been identified as necessary for normal male behavior. However, these
previously described genes have been identified using mutant screening, a process that
creates alleles with severe defects. These alleles are unlikely to persist in a natural
population, and thus the role of these genes in natural behavioral variation is unknown.
Subtle gene expression differences are more likely to be present in natural populations,
and these differences can be the target of selection. To investigate the possible role of
differential gene expression in natural variation of male courtship behavior, I identified
10 families with High Male Courtship Progression (MCP) and 10 families with Low
MCP. The full-genome expression profiles of the High and Low categories were
compared using microarrays. Fourteen candidate genes were identified as being
differentially expressed, and none of them had previously been associated with male
courtship behavior in D. melanogaster. Their functions and potential roles in courtship
are discussed. Genes contributing to natural variation in male behavior may not be the
ones previously identified in mutant screening.
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Introduction
A fundamental goal in the field of behavior genetics is to understand the
underlying genetic differences contributing to natural variation in behaviors that are
crucial to fitness (Sokolowski 1998). If individual fitness depends on behavior, there
would presumably be strong directional selection minimizing variation in such traits.
However, there have been many studies identifying variation in behavioral traits (Benus
et al. 1991; Petrie and Kempenaers 1998; Anholt 2004; Sisodia and Singh 2005).
Identifying the genetic differences contributing to behavioral variation can help
determine what might be maintaining this variation and what potential a behavioral trait
has to evolve.
Much progress has been made in our understanding of the genes underlying
courtship behavior using the fruit fly, Drosophila melanogaster, as a model system
(Sokolowski 1998; Greenspan and Dierick 2004). Male courtship behavior consists of a
complex pattern of behaviors that generally follows the sequence of orienting toward the
female, tapping her abdomen, following her, performing a courtship “song”, licking her
genitalia, attempting copulation, and copulation (Hall 1994a). Approximately 45 genes
have been identified as being involved in male courtship behavior (Drysdale et al. 2005);
these courtship deficient genes affect a range of factors, including chemosensation,
vision, courtship song production, and overall vigor and drive (Hall 1994a).
The majority of courtship deficient genes have been identified using mutant
screening (Hall 1994a; Yamamoto and Nakano 1998; Greenspan and Ferveur 2000;
Emmons and Lipton 2003). Mutant screening is useful in identifying genes involved in
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. normal courtship behavior. However, this method usually produces null alleles usually
have many pleiotropic in addition to the disruption of normal courtship behavior, and
alleles with such severe mutations are unlikely to persist at an appreciable frequency in
natural populations (Hall 1994b; Greenspan 1997). Thus it would be beneficial to
examine subtler changes in the underlying genetic architecture of natural behavioral
variation to identify potential targets of selection (Greenspan 1997; Sokolowski 1998;
Emmons and Lipton 2003). One such subtle change is gene expression; evolution via
gene expression and mutations in regulatory regions has been proposed as a major
i contributor to phenotypic variation between species (King and Wilson 1975).
Differences in gene expression both between and within species show evidence of
positive and stabilizing selection (Ferea et al. 1999; Oleksiak et al. 2002; Rifkin et al.
2003; but see Khaitovich et al. 2004). Substantial variation in expression has been
identified within populations (Oleksiak et al. 2002; Wayne et al. 2004; Hughes et al.
2006). Therefore, changes in gene expression could be an underlying cause of behavioral
variation. Static baseline expression differences can be targets of selection (e.g.
Fitzpatrick et al. 2007). Additionally, baseline expression of genes could be an important
contributor to variation in male courtship behavior. A screen of changing expression of
previously identified courtship deficient genes showed no differential expression due to
experience with females (E. A. Ruedi, unpublished data). Thus, examining the
correlation of baseline expression differences and behavioral differences of mature males
within a natural population will help identify which genes are potentially contributing to
behavioral variation.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In this study, I specifically wanted to identify genes contributing to natural
variation in D. melanogaster male courtship progression, or patterns of behavior over
time. Microarray analysis allowed me to scan the entire genome for candidate genes
associated with natural variation, as opposed to limiting my focus to previously identified
genes. I found that the genes originally considered previously associated with courtship
through to mutant analysis do not have significant differential expression associated with
high or low male courtship progression, and identified 14 new candidates that may be
related to natural variation in this behavior.
M ethods
Experimental organisms
For details about the fly stocks, mating design, and behavioral methods used here,
please refer to Chapter 2. To investigate genetic variation in male courtship behavior, I
created 80 full-sibling families from single-pair matings from the New Jersey stock.
These families were produced in four blocks of 20 families, and one block of 20 was
analyzed per week. To produce a full-sibling family, a virgin male and female were
placed on fresh vial of media, allowed to reproduce for four days, and then transferred to
another fresh vial for four more days. This reduced variation in behavior due to
environmental rearing, as full-sibling brothers from both vials were used in courtship
trials. Virgin males were collected from these vials 0-4 hours post-eclosion, and were
kept in fresh vials separated by family and date of emergence. 5-8 of these virgin
brothers were used in courtship trials when they were 5-6 days old.
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Females used as courtship targets were collected from a stock in which a
spontaneously-occurring ebony (e) mutation had been introgressed into a non-inbred wild
type genetic background (Ives background). The e/e females used in the behavioral trials
were collected as virgins 0-4 hours post-eclosion and kept in single-sex vials separated by
date of emergence.
Behavioral observations
The day before a courtship trial, males and females were placed in observation
devices modeled after those in Drapeau and Long (2000). Four total devices were set up
per day, and there were four days of trials per week (total of 16 trials/week). A total of 2
males from each family were observed each day. All trials within one week were
considered a “block”, and effects of block were included in my microarray analysis (see
below).
All behavioral observations took place between 8 am and 10:30 am and were
conducted at 24°C ± 1°C, in a room isolated from all other activities. At the start of
observations, the opaque plastic sheet separating the sexes was removed; thus a single
male was exposed to a single female. The behavior of each male was then recorded via
instantaneous scan sampling (Martin and Bateson 1993) every 30 seconds for a total of
30 minutes. At each scan, the male was scored as performing one of the behaviors in an
ordinal scale modeled after the natural male courtship progression (Male Courtship
Progression, MCP; Table 4.1). At the end of a trial, each male was thus characterized by
60 scores in a temporal sequence. All observations were conducted by the same observer
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (E.A.R.), and were blind with respect to the genotype of the male. After the behavioral
trials, all flies were discarded.
Identification of “High” and “Low” MCP families
Genetic variation in male courtship pattern was analyzed using a generalized
linear model with repeated measures, and a cumulative logit link appropriate for ordinal
data with a multinomial distribution (PROC GENMOD, SAS Statistical Analysis
Software v. 9.1; (SAS Institute 2006). Only the first 30 time points (the first 15 minutes)
were used for this analysis, as males generally cease locomotor activity towards the end
of the trial. I identified significant genetic variation in the temporal pattern of behavior
between families (family*time: x2 79=105.12, p = 0.026). Subsequently I wished to
identify 10 “High MCP” and 10 “Low MCP” families to use in gene expression analysis.
To accomplish this, I created a summary index of MCP for each individual by calculating
the rank order for each male at each observation, then calculating the mean rank for the
individual over the first 30 time intervals in the mating trial (SAS PROC RANK). Due to
the nature of the ordinal scale used in this analysis (l=motionless, 8=copulation), a low
rank correlated to a high MCP score. At each time point, tied ranks were assigned the
average rank for the tied males. Individual mean rank was then used as the dependent
variable in a linear model (SAS PROC MIXED) with family as a random effect, and day
and trial within day as fixed effects. A similar model treating family as a fixed effect was
also used to calculate the least-square mean rank for each family in an experiment. There
was significant genetic variance in rank of MCP (Log likelihood ratio test: y?\=l,
p=0.008). I identified the ten families with the highest average MCP/lowest average rank
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and the ten families with the lowest average MCP/highest average rank (Table 4.2; Figure
4.1) to use in the following microarray analysis of gene expression. These will be further
referred to as the “High” and “Low” categories.
Microarray hybridization
Tissue for RNA analysis was collected from full-sibling brothers of males used in
behavioral trials. During each block of behavioral trials, 4 males from each family were
frozen; 2 from the first single-pair mating vial, and 2 from the second single-pair mating
vial. Males from both vials were pooled into one sample per family in subsequent gene
expression analyses. All males were frozen between 2-3 pm. The frozen males had not
experienced courtship or copulation, but were the same age as the males used in courtship
observations. Expression differences due to courtship experience will therefore not be
represented in this analysis. However, an examination of dynamic expression in genes
previously associated with courtship behavior revealed that expression did not change
due to courtship experience (E. A. Ruedi, unpublished data).
I removed the heads of the four individuals frozen from each family and pooled
them together into one biological replicate to yield sufficient RNA for analysis. Heads
were used as opposed to whole flies because I was specifically interested in expression
differences for genes involved in nervous system function. RNA was extracted using the
PicoPure RNA extraction kit (Arcturus) following the manufacturer’s protocol for
extraction using DNaseI (Qiagen) treatment. Quantity and purity of RNA was checked
using a spectrophotometer (Nanodrop); the A 260/A280 for each sample was 1.9 or higher.
Quality was also checked using an Agilent 2100 bioanalyzer chip. mRNA from each
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sample was then labeled and fragmented using the Affymetrix GeneChip® One-Cycle
Target Labeling and Control Reagents kit and manufacturers instructions. Quality of
biotinylated cRNA was checked before and after fragmentation using gel electrophoresis.
Labeling was completed in two different sets that were two days apart. Samples from
both High and Low categories were equally represented in each set. Individual samples
were hybridized to Affymetrix GeneChip® Drosophila Genome 2.0 Arrays and scanned
by the University of Illinois at Urbana-Champaign Keck Center for Functional Genomics.
This provided me with 20 total gene expression profiles, 1 per family, 10 profiles per
category.
Microarray statistics
Probe hybridization intensities underwent MAS5 present/absent calls using the
“mas5calls” function (Affy package) in R Statistical Software v. 2.5 (R Development
Core Team, 2007). GCRMA normalization, transformation, and summation was then
completed using the “gcrma” function (Affy package) for the 12,148 genes called
“present.” The normalized expression levels for each gene were then analyzed using a
mixed model ANOVA (SAS PROC MIXED) with category, block, and set as fixed
effects, and block*category as a random interaction effect. The “block” effect referred to
the week that the family’s behavior was observed; the “set” effect referred to the date that
the RNA was labeled and fragmented. Some blocks did not contain both High and Low
categories, and thus the simple effect of category or the interaction effect of
block*category was not appropriate for calculating the ANOVA test statistics. Therefore,
to estimate expression differences between High and Low categories taking variance due
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to block into account, I used the “estimate” statement in SAS PROC MIXED to calculate
the correct test statistic for each gene. The “estimate” statement allowed me to control
which cat*block effects were included in the estimation of differences between High and
Low categories. To investigate the significance of expression differences for each gene, I
ran 6000 permutations of the data and analyzed those permutations using the same
statistical model and “estimate” statements as above. This allowed me to identify how
many permutations randomly produced test statistics larger than the observed data. I also
calculated the fold-change differences for any significant genes by calculating the ratio of
average expression of H/L categories (these data were not log 2 -transformed).
To examine the ability of the expression profiles from the significant genes to
appropriately classify flies into High and Low categories, I performed a linear
discriminant analysis (Everitt 2005) with the “Ida” function in R statistical software v. 2.5
(MASS package). I then used the “predict” function to classify individual flies based on
leave-one-out cross-validation of the classification using the “CV” command in the “Ida”
function.
To evaluate which biological functions were specifically enriched, and to identify
functionally redundant annotations within my candidates, I used the Gene Ontology
Functional Annotation Clustering tool in DAVID (Dennis et al. 2003). I used
PathwayAssist 3.0 to identify any gene networks associated with my candidate genes, and
the Ingenuity Pathway Analysis 5.0 software to identify any human/mouse/rat homologs
to my candidates and their corresponding pathways.
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R esults
Fourteen genes were identified as candidates contributing to natural variation in
male courtship progression. These genes were differentially expressed between High and
Low families with a probability of false positive of <0.0004 according to the permutation
tests and an FDR of <0.3. Using these cutoffs, four of the fourteen genes are likely false
positives. The names/IDs, locations, known or inferred molecular functions
(flybase.bio.indiana.edu), fold-changes and estimates of differential expression for all
fourteen genes are listed in Table 4.3. Cross-validation of the linear discriminant
analysis for these genes revealed that the expression profiles are very successful at
predicting the category of individuals (High: 90%; Low: 90%). Overall, more genes had
increased expression (13) than decreased expression (1) in the High categories compared
to the Low categories; this trend was significant (x i = 10.29, p < 0.01).
DAVID functional analysis clustering revealed two groups of enriched or
overlapping biological processes among the candidates. Functional group 1 included
cellular, macromolecule, and protein metabolism (enrichment score: 0.99, geometric
group mean P-value = 0.013; genes: pip, CG1531, Tpr2,1(3)1X-14, CG31619, CG10444,
mus312, CG9243)\ functional group 2 included the parent group of metabolism, cellular
and physiological processes (enrichment score: 0.97, geometric group mean P-value=
0.03; genes: pip,jagn, CG9243, CGI531, GluCla, mus312, Tpr2,1(3)1X-14, CG31619,
CG1819, CG10444).
PathwayAssist did not identify any known pathways associated with my candidate
genes. Ingenuity Pathways Analysis identified one pathway in which the mammalian
homolog of pip [prolactin-induced protein ) is involved. The top functions of this
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pathway are cellular development, hematological system development and function, and
immune and lymphatic system development and function. Other genes in this pathway
are early growth response 2 {Drosophila homolog: Krox-20), spleen focus forming virus
proviral integration oncogene spil, transcription factor 3, and CD4 molecule. Their
pathway network is displayed in Figure 4.2.
Discussion
Fourteen candidate genes were identified in this genome-wide analysis, and none
of them have been previously associated with male courtship behavior. At least in this
population, the expression differences of candidate courtship genes discovered using
mutation screening do not appear to be directly involved in natural behavioral variation.
This could indicate that mutation screening is useful in identifying genes required for the
development or maintenance of normal male behavior, it might not be identifying
possible targets for selection. However, I do not know where the polymorphisms causing
these expression differences are located. Thus, I cannot rule out the possibility that the
genes previously identified through mutant screening have a regulatory role in expression
differences of genes identified in this microarray study.
Other microarray analyses of expression differences contributing to female
mating speed (Mackay et al. 2005) and male aggression (Dierick and Greenspan 2006;
Edwards et al. 2006) identified a much larger number of candidate genes than the present
study. These analyses of behavior focused on lines that were artificially selected to
promote extreme differences in phenotype, whereas my analysis used standing variation
for a behavioral trait. This standing variation is much more subtle; additionally “High
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MCP families” (or “Low MCP families”) could be “High” (or “Low”) through different
means. For example, one family might display “High” MCP scores because of their
speedy response to stimuli, while another family might be “High” due to the length of
their courtship song. These differences might be attributed to different molecular
pathways, and could be the cause for such discrepancies in the number of candidate genes
between selection studies and the present study.
The functional annotation clustering revealed two main important biological
functions enriched in my candidate genes: cellular processes and metabolism. Several of
the candidate genes involved in cellular processes have possible roles in neurotransmitter
signaling; for example, GluCla is a glutamate-gated chloride channel (Cully et al. 1996;
Semenov and Pak 1999) and mutants for this gene show a similarity to other bang-
sensitive mutants previously associated with male courtship song (Ma et al. 2001). Other
candidates, such as CG10444 and CG31619, are also putatively involved in cellular
signaling in the brain. A conserved protein domain BLAST search for CG9243, the only
candidate with down-regulated expression in High families, revealed a conserved
Phospholipase D motif, which can be involved in signal transduction
(www.ncbi.nlm.nih.gov). These results indicate that variation in neurotransmission
could lead to variation in male courtship behavior, possibly by altering the
responsiveness of the fly to stimuli from the female.
The other functional cluster involved metabolic processes. These can range from
homeostasis to digestion to inhibition of hormones,pipe, the candidate with the largest
fold-change between High and Low families, is a sulfotransferase presumably involved in
metabolism. One of its 10 isoforms, ST-2, is well documented due to its involvement in
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. establishing embryonic dorsoventral axes during development through expression in the
follicle membrane (Sergeev et al. 2001). Other isoforms are also expressed in larval
salivary glands (Zhu et al. 2005), and the target of its sulfation is currently unknown.
Thepipe probe set identified in my study is specific to the ST-1 isoform; hence, pipe has
an important function in adult males that has not yet been explored.
Like pipe, many of these candidate genes have been previously associated with
development. Tpr2 is a putative protein suppressor that is important for the development
of the eye (Kazemi-Esfarjani and Benzer 2000). jagunal is an endoplasmic reticulum
(ER) protein that works to condense the ER in the oocyte during oogenesis, which allows
an increase in membrane traffic and appropriate oocyte and bristle growth (Lee and
Cooley 2007). CGI 0444 and CG31619 are both involved in cell communication
important to embryonic development and cell differentiation (Orian et al. 2003; Stark et
al. 2003; Vogel et al. 2003). The role of these genes in adults has also never been
evaluated.
Several of these genes show evidence of behavioral pleiotropy, and warrant a
much more extensive investigation into their role in male courtship progression. CG1531,
CG30492, and CG31619 all show significant differential expression between females
from lines selected for fast and slow mating speed (Mackay et al. 2005); there is
potentially a role for these genes in both male and female mating behavior. CG30492 is
also up-regulated in males from lines of flies selected for increased aggressiveness, and
thus could be involved in many behavioral processes (Dierick and Greenspan 2006).
Another candidate gene from this study, CGI 0444, is one of the down-regulated in these
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. highly aggressive lines; down-regulation in the aggressive lines was significantly less
common than up-regulation (Dierick and Greenspan 2006).
Much of the variation between the behavioral patterns of the High and Low lines
appears to occur early in the courtship trial (see Figure 4.1). This early behavioral
variation could potentially correspond to variation in the ability of different families to
detect the presence of a female and respond appropriately in a timely manner. The
candidate genes identified here indicate that cell-cell communication and metabolic
processes in the brain are important factors contributing to natural variation. Perhaps
variation in the expression of these genes corresponds to variation in communicating the
presence of stimuli and triggering appropriate behavioral responses.
This study brings us one step closer to understanding the genetic basis of natural
variation in behavior. Additional studies must be conducted to investigate the role of
expression differences in the fourteen new candidate genes described here in the
evolution of this trait, and the maintenance of variation in male courtship progression.
First, further work must be done over- or under-expressing these candidates and
examining behavioral changes in the flies to further confirm their role in natural
behavioral variation. Future studies could also investigate the presence of single
nucleotide polymorphisms (SNPs) in these new candidates to determine if expression
differences are due to polymorphisms within the coding region or perhaps in regulatory
regions. If there are no SNPs in the coding regions of the new candidates, it would be
interesting to examine sequence differences in the candidates previously identified
through mutant screening to determine if they perhaps serve a regulatory role in
differential gene expression involved in natural behavioral variation.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements
I would like to thank the National Science Foundation for their funding for this project
(NSF DEB 0296177), as well as the University of Illinois Program in Ecology and
Evolutionary Biology. I would also like to thank the members of the Hughes lab for their
help with the flies. I would especially like to thank J. M. Dmevich for her invaluable
help with microarray techniques, and T. M. Felix for learning them with me. Thanks to
the UIUC W. M. Keck Center for Functional Genomics and C. A. Wilson for their advice
and assistance. Finally, 1 would like to thank R. M. Reynolds and A. L. Toth for
reviewing this paper.
List of References
Anholt RRH (2004) Genetic modules and networks for behavior: Lessons from Drosophila. Bioessays 26(12): 1299-1306. Benus RF, Bohus B, Koolhaas JM, Vanoortmerssen GA (1991) Heritable variation for aggression as a reflection of individual coping strategies. Experientia 47(10): 1008-1019. Cully DF, Paress PS, Liu KK, Schaeffer JM, Arena JP (1996) Identification of a
Drosophila melanogaster glutamate-gated chloride channel sensitive to the antiparasitic agent avermectin. Journal o f Biological Chemistry 271(33): 20187-20191. Dennis GJ, Sherman BT, Hosack DA, Yang J, Gao W et al. (2003) David: Database for annotation, visualization, and integrated discovery. Genome Biology 4(5): P3. Dierick HA, Greenspan RJ (2006) Molecular analysis of flies selected for aggressive behavior. Nature Genetics 38(9): 1023-1031. Drapeau MD, Long AD (2000) The Copulatron, a multi-chamber apparatus for observing drosophila courtship behaviors. Drosophila Information Services 83: 194-196.
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Drysdale R, Crosby M, Consortium TF (2005) Flybase: Genes and gene models. Nucleic Acids Research 33: D390-D395 http://flvbase.org/. Edwards AC, Rollmann SM, Morgan TJ, Mackay TFC (2006) Quantitative genomics of aggressive behavior in Drosophila melanogaster. Plos Genetics 2(9): 1386-1395. Emmons SW, Lipton J (2003) Genetic basis of male sexual behavior. J Neurobiol 54(1): 93-110. Everitt B (2005) An R and S-Plus Companion to Multivariate Analysis; Casella G, Fienberg S, Olkin I, editors. London: Springer-Verlag London Limited. Ferea TL, Botstein D, Brown PO, Rosenzweig RF (1999) Systematic changes in gene expression patterns following adaptive evolution in yeast. Proceedings of the National Academy o f Sciences o f the United States o f America 96(17): 9721-9726. Fitzpatrick MJ, Feder E, Rowe L, Sokolowski MB (2007) Maintaining a behaviour polymorphism by frequency-dependent selection on a single gene. Nature 447(7141): 210-U215. Greenspan RJ (1997) A kinder, gentler genetic analysis of behavior: Dissection gives way to modulation. Curr Opin Neurobiol 7(6): 805-811. Greenspan RJ, Ferveur JF (2000) Courtship in Drosophila. AnnuRev Genet 34: 205-232. Greenspan RJ, Dierick HA (2004) 'Am not I a fly like thee?' - from genes in fruit flies to behavior in humans. Hum Mol Genet 13: R267-R273. Hall JC (1994a) The mating of a fly. Science 264(5166): 1702-1714. Hall JC (1994b) Pleiotropy o f Behavioral Genes. In: Greenspan RJ, Kyriacou CP, editors. Flexability and constraint in behavioral systems: John Wiley & Sons, LTD. pp. 15-27. Hughes KA, Ayroles JF, Reedy MM, Drnevich JM, Rowe KC et al. (2006) Segregating variation in the transcriptome: cis regulation and additivity of effects. Genetics 173(3): 1347-1364. Kazemi-Esfarjani P, Benzer S (2000) Genetic suppression of polyglutamine toxicity in Drosophila. Science 287(5459): 1837-1840. Khaitovich P, Weiss G, Lachmann M, Hellmann I, Enard W et al. (2004) A neutral model of transcriptome evolution. Plos Biology 2(5): 682-689.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. King M-C, Wilson AC (1975) Evolution at two levels in humans and chimpanzees. Science 188(4184): 107-116. Lee S, Cooley L (2007) jagunal is required for reorganizing the endoplasmic reticulum during Drosophila oogenesis. J Cell Biol 176(7): 941-952. Ma E, Gu XQ, Wu XH, Xu T, Haddad GG (2001) Mutation in pre-mRNA adenosine deaminase markedly attenuates neuronal tolerance to o-2 deprivation in Drosophila melanogaster. Journal o f Clinical Investigation 107(6): 685-693. Mackay TFC, Heinsohn SL, Lyman RF, Moehring AJ, Morgan TJ et al. (2005) Genetics and genomics of Drosophila mating behavior. Proceedings of the National Academy of Sciences of the United States of America 102: 6622-6629. Martin P, Bateson P (1993) Measuring Behaviour: An introductory guide. Cambridge: Cambridge University Press. Oleksiak MF, Churchill GA, Crawford DL (2002) Variation in gene expression within and among natural populations. Nature Genetics 32(2): 261-266. Orian A, van Steensel B, Delrow J, Bussemaker HJ, Li L et al. (2003) Genomic binding by theDrosophila MYC, MAX, MAD/MNT transcription factor network. Gene Dev 17(9): 1101-1114. Petrie M, Kempenaers B (1998) Extra-pair paternity in birds: Explaining variation between species and populations. Trends EcolEvol 13(2): 52-58.
Rifkin SA, Kim J, White KP (2003) Evolution of gene expression in the Drosophila melanogaster subgroup. Nature Genetics 33(2): 138-144. SAS Institute (2006) SAS online documentation v. 9.1. Available: http://support.sas.com/91doc/docMainpage.isp. Accessed 2006. Semenov EP, Pak WL (1999) Diversification of Drosophila chloride channel gene by multiple posttranscriptional mRNA modifications. J Neurochem 72(1): 66-72. Sergeev P, Streit A, Heller A, Steinmann-Zwicky M (2001) The Drosophila dorsoventral determinant pipe contains ten copies of a variable domain homologous to mammalian heparan sulfate 2-sulfotransferase. Developmental Dynamics 220(2): 122-132. Sisodia S, Singh BN (2005) Behaviour genetics of Drosophila: Non-sexual behaviour. Journal of Genetics 84(2): 195-216.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sokolowski MB (1998) Genes for normal behavioral variation: Recent clues from flies and worms. Neuron 21(3): 463-466. Stark A, Brennecke J, Russell RB, Cohen SM (2003) Identification of Drosophila microRNA targets. Plos Biology 1(3): 397-409. Vogel C, Teichmann SA, Chothia C (2003) The immunoglobulin superfamily in Drosophila melanogaster and Caenorhabditis elegans and the evolution of complexity. Development 130(25): 6317-6328. Wayne ML, Pan YJ, Nuzhdin SV, McIntyre LM (2004) Additivity and trans- acting effects on gene expression in male Drosophila simulans. Genetics 168(3): 1413-1420. Yamamoto D, Nakano Y (1998) Genes for sexual behavior. Biochemical & Biophysical Research Communications 246(1): 1-6. Zhu XJ, Sen J, Stevens L, Goltz JS, Stein D (2005) Drosophila pipe protein activity in the ovary and the embryonic salivary gland does not require heparan sulfate glycosaminoglycans. Development 132(17): 3813-3822.
I l l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLES
Table 4.1. Male Courtship Progression Scale
Score Behavior
1 Motionless
2 M oving
3 Oriented toward female
4 Following/chasing female
5 Wing vibration
6 Licking genitalia
7 Attempted copulation
8 Copulation
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.2. Average rank and average MCP score for 10 High and 10 Low families.
Family1 Rank2 ± Std. err. Average MCP ± Std. err.
30 380.27 ± 38.88 7.22 ±0.17
153 376.28 ±41.57 7.25 ±0.19
61 373.26 ±38.88 7.23 ±0.17
167 362.60 ±38.88 6.99 ±0.17
113 362.30 ±49.18 7.02 ± 0.22 95 358.60 ±38.88 6.93 ±0.17 57 352.18 ±41.57 6.81 ±0.19
92 352.17 ±38.88 6.87 ±0.17
139 347.37 ± 38.88 6.94 ±0.17
125 344.86 ±41.57 6.96 ±0.19
177 239.18 ±38.88 4.85 ±0.17
12 235.20 ±44.90 4.71 ±0.20 54 233.50 ±38.88 4.89 ±0.17
21 220.80 ±38.88 4.55 ±0.17
32 216.05 ±41.57 4.20 ±0.19
182 203.04 ±41.57 3.90 ±0.19
65 193.22 ±38.88 3.83 ±0.17
15 190.11 ±41.57 3.84 ±0.19
35 170.71 ±38.88 3.28 ±0.17
87 148.79 ±41.57 2.84 ±0.19
1- Families in bold are "High MCP"; families in italic are “Low MCP” 2- Lower ranking indicates higher MCP score
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0 0.0002 0.0002 0.0002 0.0002 0.04 0.0002 0.04 0.04 0.0002 5.29 0.03 0 4.6 0.46 0.31 0.24 4.59 0.34 4.92 0.04 Effect Size2 t-Value3 Prob-t PFP4
1.46 0.551.24 4.65 0.311.24 4.19 0.05 0 1.42 0.511.18 5.43 0.03 1.27 1.31 0.39 4.54 0.05 (known or inferred) lipase and phospholipase activity -1.26 -0.33 -4.43 0.05 glutamate-gated Cl channel activity; zinc ion binding cholinesterase activity Na-dep. multivitamin transporter act. unknown unknown neurotransmitter receptor activity 3R85F14-85F15 metalloendopeptidase activity; 2L39F1-39F32R 43E5-43E7 procollagen N-endopeptidase act. zinc ion binding 1.49 0.57 4.36 0.05 0 3L 3L 65F4-65F5 protein binding 1.37 3R92B1-92B2 3R 84D8-84D9 carboxylesterase activity; 2R56F11-56F11 cation transporter activity; X X 7C9-7D1 2L 39A7-39A7 X 16B6-16B7 4.3. Genes with significantly different expression between High MCP and Low MCP categories a-EstlO CG30492 CG31619 1(3)1X-14 C G I0444 CG1531 CG9243 GluCla CG12991 mus312 Gene Name/ID Location1 Molecular Function Fold Change Table
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.0003 0.0003 0.05 0.05 4.4 4.59 0.04 0.0003 4.55 0.19 3.23 1.14 1.25 0.32 Fold Change Effect Size2 t-Value3 Prob-t PFP4
(known or inferred) heat shock protein bindingATP binding; transmembranemovement; transporter activity 1.21 0.28 4.16 0.05 0.0003 Location1 Molecular Function Genes with significantly different expression between High MCP and Low MCP categories 3L 76A5-76A6 sulfotransferase activity 9.38 3R 83C3-83C3 unknown 2L 36A2-36A2 X N/A (Con’t). 4.3 1 cytologicalChromosome and 1 arm position Tpr2 CG34120 Gene Name/ID with cutoffa of0.0004,< PFP approximately 4 genes out of this list can be expected to be false positive. 3 3 Degrees o = 2 ffreedom for all t-tests Table 2 Average difference ofmRNA abundance in High vs. Low categories on Log2 scale 4 of Probability significantdetecting difference in mRNA expression between High v. Low categories due to chance (out of6000 permutations ofdata); pip jagn
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced Figure 4.1. 4.1. Figure over the first 15 m inutes o f the courtship trial. courtship the f o inutes m 15 first the over I S E R U FIG Average MCP Score M ale courtship progression o f f o progression courtship ale M Time (Minutes) Time 10 116 “High M CP” and and CP” M “High 10 “Low M CP” fam ilies ilies fam CP” M “Low Families Figure 4.2. Pathway of mammalian homolog(proactin-involvedprotein) topipe.
SPI1
T C F 3 ^
\ / " EGR2 \ / V'/C / * / I \ ' / kCD4y
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 5: Conclusion
Natural variation in the progression of male courtship behavior in Drosophila
melanogaster could be the target of selection. Although many genes have been
implicated in contributing to normal male courtship behavior in this species, it is not
known if these are the same genes contributing to natural variation. The goals of my
dissertation were to identify variation in the sequence of male behavior using a natural
population, and investigate the genetic underpinnings of this variation. The latter was
accomplished by examining alterations in expression of previously identified candidate
genes in response to changes in age or experience, and then investigating if the
previously identified candidates were also involved in natural behavioral variation. My
thesis had the following objectives:
Objective 1: Develop a high-throughput assay of male courtship behavior that captures the details of courtship progression for a sample size large enough for genetic and genomic studies. Use this male courtship progression (MCP) assay to determine the presence of genetic variation in the whole sequence of courtship behavior within a natural population.
Genetic differences in MCP were detected whether males were placed with virgin
or previously mated female targets. The new MCP assay I developed was the only metric
able to detect variation in male behavior directed towards mated females, arguably the
type of female most often encountered in the wild. Variation in the sequence of male
courtship has a genetic basis, and thus could be the target of selection. I then wanted to
investigate the possible molecular contributions to natural variation. I focused on
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. differential gene expression, as alteration of the expression of a gene can be one target of
selection. Part of this alteration can be due to non-genotypic factors, such as age or
experience.
Objective 2: Examine candidate courtship genes previously identified in the literature for alteration in gene expression due to age or experience with a female. Genes were considered static if age, experience, or the interaction between age and experience had no effect on expression. Genes were considered dynamically expressed if any of these factors had a significant effect on expression.
In the set of previously identified courtship genes, I found that any dynamic
changes in gene expression occur due to age (reproductive maturity), not experience with
females. There was a marginally non-significant effect of the interaction between age
and experience, which hints at the possibility of an expression*environment interaction.
Future work could examine the full-genome expression profiles for more replicates of
isogenic males of different ages and experience levels. This would confirm the results of
my dissertation, as well as identify any new genes responding to experience with a
female.
I continued my investigation of genes contributing to natural variation in male
courtship progression using mature males. I wanted to determine which genes exhibited
differential baseline expression between families with the most extreme differences in
courtship progression from Objective 1.
Objective 3: Use full-genome microarray analysis to find candidate genes associated with High and Low male courtship progression. Compare these candidates with
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. previously identified candidates, and examine functional trends in the new set of candidates.
I discovered 14 candidate genes with expression differences correlated to High
and Low MCP; none were previously associated with male courtship behavior. I also
identified which of these new candidates might be co-regulated. Many of the new
candidates have been implicated in cellular communication and development. I suggest
that variation in the expression of these genes could correspond to variation in
communicating the presence of stimuli and triggering appropriate behavioral responses.
Future research will confirm that these candidates are directly associated with male
courtship progression by using functional mutants with decreased (or silenced)
expression to examine their effects on behavior.
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix A: Statistical tests for univariate models of single behaviors
E ffect1
Family Day Trial(Day)
Trait
VT Experiment
Motionless 1.78(77,409) * 1.27(14,409) 1.04 (48,409)
Moving 1.68 (77,409) * 2.09(14,409) 1.56(48,409)
Orienting 1.38 (77,409) 1.04(14,409) 1.17(48,409)
Follow ing 0.89 (77,409) 1.07(14,409) 0.78 (48,409)
Wing Vibration 1.09 (77,409) 0.78 (14,409) 0.85 (48,409)
Licking 0.86 (77,409) 1.15 (14,409) 1.21 (48,409)
Attempted Copulation 1.28 (77,409) 1.23 (14,409) 1.51 (48,409)
M T Experiment
Motionless 1.21 (75,374) 1.53 (12,374) 0.97 (48,374)
Moving 1.10(75,374) 1.28 (12,374) 1.85 (48,374)**
Orienting 1.18 (75,374) 1.33 (12,374) 1.08 (48,374)
Follow ing 1.10(75,374) 0.91 (12,374) 0.87 (48,374)
Wing Vibration 0.97 (75,374) 1.02(12,374) 0.76 (48,374)
Licking 1.08 (75,374) 2.08 (12,374) 1.33 (48,374)
Attempted Copulation 1.22 (75,374) 0.67 (12,374) 1.04 (48,374)
* p<0.05; ** p<0.01; *** p<0.001; Sequential Bonferonni used to correct for multiple tests
1 - Family, day, and trail(day) tested with Type III F-statistics
121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix B: Mean MCP scores for all VT families
Family Average MCP Score Standard Error
3 5.921 0.174
6 6.590 0.186
7 5.075 0.174
8 6.489 0.201
12 4.711 0.201
14 4.696 0.174
15 3.838 0.186
16 5.779 0.174
17 5.775 0.174
18 5.362 0.186
19 5.689 0.201
21 4.546 0.174
26 6.239 0.201
28 5.362 0.186
29 5.911 0.201
30 7.217 0.174
32 4.205 0.186
35 3.275 0.174
38 5.510 0.186
39 7.100 0.220
54 4.892 0.174
57 6.810 0.186
61 7.233 0.174 62 6.750 0.174
64 6.752 0.186
65 3.825 0.174
68 5.575 0.174
72 6.305 0.186
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix B (c o n ’t)
Fam ily Average MCP Score Standard Error 74 6.029 0.174 75 5.908 0.174 77 4.814 0.186 78 6.029 0.186 79 6.500 0.174 82 6.533 0.174 83 6.058 0.174 84 6.690 0.186 87 2.843 0.186 88 '5.567 0.174 92 6.871 0.174 95 6.933 0.174 101 5.395 0.186 103 5.605 0.186 107 5.842 0.174 108 6.163 0.174 109 5.961 0.201 112 6.086 0.186 113 7.020 0.220 114 6.546 0.174 115 5.867 0.174 117 6.446 0.174 119 5.588 0.174 120 6.243 0.186 121 5.346 0.174 124 6.088 0.174 125 6.962 0.186
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix B (c o n ’t)
Fam ily Average MCP Score Standard Error 135 6.210 0.186 136 5.054 0.174 137 6.714 0.186 139 6.938 0.174 146 4.208 0.174 152 5.846 0.174 153 7.248 0.186 156 6.319 0.186 157 6.117 0.174 159 5.181 0.186 160 6.763 0.174 161 6.396 0.174 163 6.417 0.174 167 6.988 0.174 169 6.519 0.186 170 6.083 0.174
174 6.071 0.174 176 6.810 0.186 177 4.854 0.174 182 3.900 0.186 183 6.067 0.201 184 5.786 0.186 186 5.575 0.174 189 6.646 0.174 193 5.733 0.174
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix C: Mean MCP scores for all MT families
F am ily Average MCP Score Standard Error
1 3.750 0.149
4 3.144 0.149
8 3.362 0.138
10 4.100 0.138
13 3.380 0.164
16 4.305 0.138
20 3.367 0.129
21 1.910 0.138
25 4.024 0.138
28 3.921 0.129
34 3.044 0.149
38 3.733 0.138
39 3.433 0.149
40 3.763 0.129
43 2.954 0.129
44 3.804 0.129
46 3.110 0.138
47 2.613 0.129
49 3.325 0.129
53 3.240 0.164
54 4.213 0.164
61 3.710 0.138
62 4.083 0.149 71 2.567 0.129
72 2.706 0.149
74 3.307 0.164
77 2.789 0.149
78 4.506 0.149
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix C (c o n ’t)
Fam ily Average MCP Score Standard Error 80 2.756 0.149 81 3.796 0.129 84 4.057 0.138 86 3.978 0.149 90 3.943 0.138 91 3.233 0.138 93 3.583 0.129 94 3.371 0.138 98 2.692 0.183 99 3.852 0.138 100 4.389 0.149 101 3.229 0.129 102 3.375 0.129 103 3.111 0.149 111 3.321 0.129 113 3.217 0.183 114 2.556 0.149 115 4.552 0.138 119 2.652 0.138 121 3.829 0.129 123 4.361 0.149 127 3.200 0.138 129 4.081 0.138 135 3.524 0.138 140 3.467 0.164 143 4.104 0.129 144 3.767 0.149 145 4.548 0.138
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix C (c o n ’t)
Fam ily Average MCP Score Standard Error 146 3.790 0.138 148 4.292 0.129 149 3.983 0.149 151 4.524 0.138 152 4.167 0.164 154 3.207 0.164 156 2.983 0.183 157 3.119 0.138 160 2.952 0.138 161 2.889 0.149 162 3.808 0.129 164 3.933 0.138 166 3.519 0.138 167 4.200 0.149 172 3.028 0.149 173 3.690 0.138
174 3.744 0.149 182 2.846 0.129 184 3.489 0.149 185 3.657 0.138 187 2.805 0.138 191 4.086 0.138 193 3.476 0.138
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D: Least square means of gene expression due to age,
experience, or age*experience.1
Age
Gene Age Estimate Std.Err.
18w I -0.469 0.050
18w M -0.482 0.054
amn I -2.016 0.044
amn M -1.946 0.048
ato I -2.354 0.032
ato M -2.347 0.033
CaMKII I 0.133 0.024
CaM KII M 0.189 0.026
cpo I -1.233 0.046
cpo M -1.095 0.050
crl I -0.200 0.031
crl M -0.155 0.034
dnc I 1.779 0.044
dnc M 1.825 0.048
dsx I -0.701 0.041
dsx M -0.525 0.043
eag I -0.266 0.032
eag M -0.164 0.035
eg I -1.765 0.044
eg M -2.208 0.047 fru I -1.296 0.036 fru M -1.111 0.038
hom er I -0.075 0.041
homer M -0.150 0.045
1 Estimates reported are log transformed
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (co n ’t) Age Gene Age Estimate Std.Err. Kr-hl I -0.444 0.039 Kr-hl M -0.286 0.041 nonA I -0.853 0.052 nonA M -0.817 0.055 para I 0.259 0.033 para M 0.361 0.036 pie I 0.667 0.036 pie M -0.154 0.040 pros I -0.048 0.029 pros M 0.091 0.031 qtc I 11.528 0.064 qtc M 11.371 0.070 rut I -0.133 0.055 rut M 0.032 0.060 Sh I 0.065 0.060 Sh M 0.083 0.065 tko I -0.956 0.048 tko M -0.764 0.052 to I -1.147 0.037 to M -0.196 0.040 tra I -0.855 0.029 tra M -0.846 0.031 tra2 I -0.988 0.072 tra2 M -1.023 0.078
y I -1.698 0.029 y M -2.088 0.032
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (co n ’t) Experience Gene Experience Estimate Std.Err. 18w N -0.439 0.050 18w E -0.512 0.054 amn N -1.987 0.044 amn E -1.975 0.048 ato N -2.341 0.032 ato E -2.360 0.033 CaMKII N 0.151 0.024 CaMKII E 0.171 0.026 cpo N -1.155 0.046 cpo E -1.172 0.050 crl N -0.199 0.031 crl E -0.156 0.034 dnc N 1.818 0.044 dnc E 1.785 0.048 dsx N -0.630 0.041 dsx E -0.596 0.043 eag N -0.215 0.032 eag E -0.216 0.035
eg N -1.961 0.044 eg E -2.013 0.047 fru N -1.175 0.037 fru E -1.232 0.037 homer N -0.096 0.041 homer E -0.129 0.045 Kr-hl E -0.329 0.041 Kr-hl N -0.401 0.039
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (c o n ’t)
Experience Gene Experience Estimate Std.Err. nonA N -0.802 0.052
nonA E -0.869 0.055 para N 0.289 0.033 para E 0.331 0.036 pie N 0.275 0.036 pie E 0.238 0.040 pros N -0.005 0.029 pros E 0.048 0.031 qtc N 11.435 0.064
qtc E 11.463 0.070 rut N -0.045 0.055 rut E -0.055 0.060 Sh N 0.133 0.060 Sh E 0.016 0.065 tko N -0.851 0.048 tko E -0.869 0.052 to N -0.723 0.037 to E -0.621 0.040 tra N -0.847 0.029 tra E -0.853 0.031 tra2 N -0.926 0.072 tra2 E -1.086 0.078
y N -1.897 0.029 y E -1.889 0.032
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (c o n ’t) Age *Experience Gene Age Experience Estimate Std.Err. 18w I N -0.456 0.070 18w I E -0.482 0.070 18w M N -0.422 0.070 18w M E -0.542 0.082 amn I N -1.963 0.063 amn I E -2.069 0.063 amn M N -2.011 0.063 amn M E -1.881 0.074 ato I N -2.315 0.047 ato I E -2.393 0.045 ato M N -2.367 0.045 ato M E -2.327 0.049 CaMKII I N 0.134 0.034 CaMKII IE 0.132 0.034 CaMKII MN 0.168 0.034
CaMKII M E 0.210 0.040 cpo I N -1.228 0.065 cpo I E -1.238 0.065 cpo M N -1.083 0.065 cpo M E -1.107 0.076 crl I N -0.219 0.044 crl I E -0.180 0.044 crl M N -0.179 0.044 crl M E -0.131 0.052
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (c o n ’t)
Age *Experience Gene Age Experience Estimate Std.Err. dnc I N 1.788 0.063 dnc I E 1.769 0.063 dnc M N 1.849 0.063 dnc M E 1.802 0.073 dsx I N -0.715 0.058 dsx I E -0.688 0.058 dsx M N -0.546 0.058 dsx M E -0.503 0.064 eag I N -0.247 0.045 eag I E -0.285 0.045 eag M N -0.182 0.045 eag M E -0.146 0.053
eg I N -1.745 0.062 eg I E -1.785 0.062 eg M N -2.176 0.062 eg M E -2.241 0.072 fru I N -1.230 0.052 fru I E -1.361 0.050 fru M N -1.121 0.052 fru M E -1.102 0.055 homer I N -0.050 0.058 homer I E -0.101 0.058 homer M N -0.143 0.058 homer M E -0.158 0.068
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (c o n ’t)
Age *Experience Gene Age Experience Estimate Std.Err. Kr-hl I N -0.426 0.056 Kr-hl I E -0.462 0.056 Kr-hl M N -0.375 0.056 Kr-hl M E -0.197 0.062 nonA I N -0.824 0.074 nonA I E -0.882 0.074 nonA M N -0.780 0.074 nonA M E -0.855 0.082 para I N 0.258 0.046 para I E 0.260 0.046 para M N 0.319 0.046 para M E 0.402 0.054 pie I N 0.677 0.051 pie I E 0.657 0.051 pie M N -0.127 0.051 pie M E -0.181 0.060 pros IN -0.056 0.041 pros I E -0.040 0.041 pros M N 0.045 0.041 pros M E 0.136 0.048 qtc I N 11.543 0.091 qtc I E 11.512 0.091
qtc M N 11.328 0.091 qtc M E 11.414 0.107
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (c o n ’t)
Age * Experience
Gene Age Experience Estimate Std. Err. rut I N -0.141 0.078 rut I E -0.125 0.078 rut M N 0.051 0.078 rut M E 0.014 0.091 Sh I N 0.141 0.085 Sh I E -0.011 0.085 Sh M N 0.124 0.085 Sh M E 0.042 0.099 tko I N -0.958 0.067 tko I E -0.954 0.067 tko M N -0.743 0.067 tko M E -0.785 0.079 to I N -1.216 0.052 to I E -1.078 0.052 to M N -0.229 0.052 to M E -0.163 0.061 tra I N -0.857 0.041 tra I E -0.852 0.041 tra MN -0.838 0.041 tra M E -0.854 0.048 tra2 I N -0.902 0.101 tra2 I E -1.075 0.101 tra2 M N -0.950 0.101 tra2 M E -1.097 0.119
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix D (c o n ’t)
Age * Experience Gene Age Experience Estimate Std. Err.
y I N -1.715 0.041 y I E -1.682 0.041 y M N -2.079 0.041 y M E -2.096 0.049
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Curriculum Vitae Elizabeth Anne Ruedi
Work Address Program in Ecology and Evolution University of Illinois at Urbana-Champaign 2414-L Institute of Genomic Biology, MC-195 1206 W. Gregory Drive Urbana, Illinois 61801 (217) 333-5887 (lab) (217)244-1781 (fax) Email: [email protected]
E d u catio n : PhD, Program in Ecology and Evolutionary Biology, University of Illinois at Urbana- Champaign—July 2007. T hesis: Natural variation and gene expression in the male courtship progression o f Drosophila melanogaster. Coursework included: Ecological Genetics; Northern Blot Techniques; Genes and Behavior; Adaptive Genetic Variation in the Wild; Applied Statistical Methods; Genomic Analysis of Insects; Design and Analysis of Biological Experiments; Interface of Ecology and Evolution; Applied Bioinformatics; Teaching College Biology
B. A. with Biology and English Majors, Lake Forest College. Lake Forest, Illinois— May 2001. Senior Thesis: Reverse Sexual Size and Shape Dimorphism in North American A ccipiter Hawks.
Publications: Hughes, K.A., Ayroles, J.F., Reedy M.M., Drnevich, J.D., Rowe, K.C., Ruedi, E.A., Caceres, C.E. and Paige, K.N. (2006). “Segregating Variation in the Transcriptome: Cis Regulation and Additivity of Effects.” Genetics 173: 1347- 1355.
Olendorf, R, Reudi, B. [s/c], and Hughes, K.A. (2004) “Primers for 12 polymorphic microsatellite DNA loci from the guppy (Poecilia reticulata).” M olecular Ecology N otes 4 (4): 668-671.
Drnevich, J. M., Reedy, M. M., Ruedi, E. A., Rodriguez-Zas, S., and Hughes, K. A. (2004). “Quantitative evolutionary genomics: Differential gene expression associated with male competitive reproductive success in Drosophila melanogaster. ” Proceedings of the Royal Society of London, Series B 271: 2267- 2273.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Publications (con’t): Ruedi, E. A., Owusu, Y. N., and Dixon, K. (2003). “Male courtship intensity: a new method of Drosophila courtship measurement.” Drosophila Information Services 86.
Presentations Given at Professional Symposia: “Natural genetic variation in the male courtship behavior of Drosophila melanogaster.” E. A. Ruedi and K. A. Hughes. Animal Behavior Society Meeting. Snowbird, Utah. August 2005.
“Evidence for Sexually Antagonistic Interactions in the Genetic Makeup of Drosophila melanogaster.” E. Ruedi. 5th Annual Graduate Student Symposium. Presented by Graduate Students in Ecology and Evolutionary Biology, University of Illinois at Urbana-Champaign. Urbana, Illinois. February 2003.
“Genetic Influence on the Courtship Behavior of Two Species of Cactophilic Drosophila: Preliminary Data and Future Research Plans.” E. Ruedi. 4lh Annual Graduate Student Symposium. Presented by Graduate Students in Ecology and Evolutionary Biology, University of Illinois at Urbana-Champaign. Urbana, Illinois. February 2002.
Posters Presented at Professional Symposia and Seminars: “Natural genetic variation in the male courtship behavior of Drosophila melanogaster .” E. A. Ruedi and K. A. Hughes. Gordon Conference: Genes and Behavior. Ventura, California. February 2006.
“Natural genetic variation in the male courtship intensity of Drosophila melanogaster.” E. A. Ruedi and K. A. Hughes. Midwest Drosophila Meeting 2004. Monticello, Illinois. October 2004.
“Genetic influence on male courtship behavior in two species of cactophilic Drosophila.” E. A. Ruedi, C. Tymczyna-Cobbs, K. A. Hughes. 9th International Behavioral Ecology Congress. Montreal, Quebec, Canada. July 2002.
“Reverse Sexual Size and Shape Dimorphism in North American Accipiter Hawks.” E. A. Ruedi and A. Houde. Evolution 2001. Knoxville, Tennessee. June 2001.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Research Experience Research Interests: Behavior, Genetics, and Evolution
Laboratory Assistant - Dr. Kimberly Hughes. University of Illinois. August 2001- Present Responsibilities: Care of Drosophila melanogaster, D. mettleri, and D. micromettleri; experimental design and execution; undergraduate employee training; data collection and analysis.
Research Assistant-Dr. Kimberly Hughes and Dr. Robert Olendorf. Summer 2003 and 2004. Responsibilities: Primer optimization; DNA extraction; running Polymerase Chain Reactions; gel electrophoresis.
Laboratory Assistant - Dr. Anne Houde. Lake Forest College. January 2001- April 2001 Responsibilities: Care of guppies ( Poecilia reticulata) for Dr. Houde’s experiments; care of Betta fish (Betta splendins ), crayfish, and house crickets (Acheta domestica) for Animal Behavior
Skills: Behavioral Observation DNA Extraction RNA Extraction PCR Agarose gel electrophoresis Northern Blot execution Reverse Transcription Quantitative real-time PCR Primer design/optimization RNA amplification and labeling for microarray analysis Experience with: Microsoft Word, Excel, PowerPoint; JMP; SAS; R.
Teaching Experience: Teaching Assistant - Integrative Biology 405, Ecological Genetics; Fall 2006 Responsibilities: Preparing and giving three 1 xh hour lectures, designing and implementing primary literature discussions, holding office hours
Teaching Assistant - Integrative Biology 150/199, Organismal and Evolutionary Biology Merit Scholars Program; Fall 2005 Responsibilities: Running a discussion section for high-achieving students planning a career in science. Focus was on group work and independent problem solving.
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Teaching Experience (con’t): Teaching Assistant - Integrative Biology 201, Genetics and Evolution; Spring 2005 Responsibilities: Running laboratory sections focusing on evolution, systematics, and genetics.
Teaching Assistant - Integrative Biology 150, Organismal and Evolutionary Biology; Fall 2003, Spring 2004, Fall 2004; Fall 2006 Responsibilities: Running four discussion sections dedicated to expanding the students’ understanding of lecture materials
Teaching Assistant - Integrative Biology 151, Laboratory on Organismal and Evolutionary Biology; Fall 2002, Spring 2003 Responsibilities: Running laboratory sections designed for non-IB majors
Peer Teacher - Ecology and Evolution, Bio 220. Dr. Anne Houde. Lake Forest College. August 2000- December 2000. Responsibilities: Orchestrating group discussions; lecture planning and lecturing; holding office hours; reviewing and editing student papers.
Teaching Honors: John G. & Evelyn Hartman Heiligenstein Outstanding Teaching Assistant in Integrative Biology 150, UIUC. Spring 2005
Incomplete List of Teachers Ranked as Excellent, University of Illinois at Urbana- Champaign—Fall 2002, Spring 2003, Fall 2003, Spring 2004, Fall 2004, Spring 2005, Fall 2006
Academic Honors: National Science Foundation Doctoral Dissertation Improvement Grant. 2006.
Magna Cum Laude. Lake Forest College Graduation 2001.
Honors in Biology. Lake Forest College, B.A. 2001.
Beta Beta Beta National Biological Honor Society. 2001 - Present.
Sigma Xi Society for Excellence in Scientific Research. 2001 - 2004.
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Service and Committee Work: Graduate Students in Ecology and Evolutionary Biology (GEEB) President—Fall 2006- Present
GEEB Vice President—Fall 2006-Spring 2006
PEEB Steering Committee Student Representative—Fall 2004- Spring 2006
GEEB Symposium Committee Member—Fall 2002- February 2003
Social Director—Fall 2002- Fall 2003
Ecolunch Coordinator—Spring 2002- Fall 2002
Reading Group Coordinator—Fall 2001-Spring 2002
Professional References: Kimberly A. Hughes, Ph.D. (Associate Professor) Department of Animal Biology University of Illinois at Urbana-Champaign 2404 Institute of Genomic Biology, me-195 1206 W. Gregory Drive Urbana, Illinois 61801 (217) 244-6632 (217) 244-1781 (fax) email: [email protected]
Gene Robinson, Ph.D. G. William Arends Professor of Integrative Biology Director, Neuroscience Program Chair, Genomics of Neural and Behavioral Plasticity Theme, Institute for Genomic Biology Department of Entomology University of Illinois 505 S. Goodwin Ave., Urbana, IL 61801 (217) 265-0309 (217) 244-3499 (fax) email: [email protected]
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Professional References (con’t): Kevin Dixon, Ph.D. (Teaching Laboratory Specialist) School of Integrative Biology University of Illinois at Urbana-Champaign 300 Natural History Building, me-120 505 South Goodwin Urbana, Illinois 61801 (217) 244- 7350 (217) 244- 1224 (fax) email: kdixon@ 1 ife.uiuc.edu
Anne E. Houde, Ph.D. (Associate Professor, Chairperson of Biology Department) Department of Biology Lake Forest College 555 N. Sheridan Road Lake Forest, Illinois 60045 (847) 735- 6043 email: [email protected]
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.