Download No Boundaries: Genomes, Organisms, and

Journal of Heredity 2014:105(Special Issue):756–770
doi:10.1093/jhered/esu039
© The American Genetic Association. 2014. All rights reserved.
For permissions, please e-mail: [email protected]
No Boundaries: Genomes, Organisms,
and Ecological Interactions Responsible
for Divergence and Reproductive
Isolation
William J. Etges
From Program in Ecology and Evolutionary Biology, Department of Biological Sciences, 1 University of Arkansas,
Fayetteville, AR.
Address correspondence to William J. Etges at the address above, or e-mail: [email protected].
Abstract
Revealing the genetic basis of traits that cause reproductive isolation, particularly premating or sexual isolation, usually involves
the same challenges as most attempts at genotype–phenotype mapping and so requires knowledge of how these traits are
expressed in different individuals, populations, and environments, particularly under natural conditions. Genetic dissection of
speciation phenotypes thus requires understanding of the internal and external contexts in which underlying genetic elements
are expressed. Gene expression is a product of complex interacting factors internal and external to the organism including
developmental programs, the genetic background including nuclear–cytotype interactions, epistatic relationships, interactions
among individuals or social effects, stochasticity, and prevailing variation in ecological conditions. Understanding of genomic
divergence associated with reproductive isolation will be facilitated by functional expression analysis of annotated genomes
in organisms with well-studied evolutionary histories, phylogenetic affinities, and known patterns of ecological variation
throughout their life cycles. I review progress and prospects for understanding the pervasive role of host plant use on genetic
and phenotypic expression of reproductive isolating mechanisms in cactophilic Drosophila mojavensis and suggest how this
system can be used as a model for revealing the genetic basis for species formation in organisms where speciation phenotypes
are under the joint influences of genetic and environmental factors.
Subject areas: Genomics and gene mapping; Gene action, regulation, and transmission
Key words: cactus, courtship song, cuticular hydrocarbon, Drosophila, gene ontology, microarray, Sonoran Desert, speciation, transcriptome
Deciphering genetic architectures of phenotypic traits associated with reproductive divergence and speciation often
requires a quantitative genetic approach to estimate variance components, gene numbers, effect size, and the extent
of genotype by environment (G × E) interactions if these
traits are polygenic and sensitive to local environmental
conditions. Study of genes with large qualitative phenotypic
effects, for example, the genetic basis of shell color and
banding patterns in species of Cepea snails, melanic morphs
of Biston betularia and other species (Majerus 1998; Cook
et al. 2012), wing color and banding patterns in Heliconius
butterflies (Turner 1977; Naisbit et al. 2003; Mallet and
Dasmahapatra 2012), and mammalian pelage color differences (Dice and Blossom 1937; Blair 1947; Kaufman 1974;
Nachman et al. 2003; Linnen et al. 2013) have remained at
the heart of ecological genetics for over 80 years largely
because one or a few gene loci are involved for each trait,
756
phenotype expression showed relatively little environmental influence, and the selective forces maintaining polymorphism were largely identifiable (Ford 1975). Some genetic
analyses of speciation have often focused on one or a few
gene regions without regard to the influences of relevant
ecological variation or controlled environmental differences
in the laboratory on reproductive isolation. For some “speciation genes,” such as those underlying species differences
involved with premating (Doi et al. 2001) and postmating
isolation (Orr et al. 2004; Ting et al. 2004), genetic analysis
has been devoid of assessing environmental influences on
these phenotypes. Fortunately, the role of ecology in driving
reproductive isolation has become better integrated into the
analysis of genetic architectures of diverging populations
and species (e.g., Bradshaw et al. 1995; Hawthorne and Via
2001; Feder et al. 2005; Egan and Funk 2009; Feder and
Nosil 2009; Gompert et al. 2014).
Etges • Genetics, Environments, and Speciation
Most traits associated with reproductive isolation, whether
premating, postmating-prezygotic, or postzygotic, are genetically complex traits and the identifying genes underlying
them has remained elusive except in a small number of
cases (Coyne and Orr 2004). Further, many kinds of phenotypes that mediate reproductive isolation are influenced by
the environments in which they are expressed, for example,
insect and amphibian courtship songs (Gerhardt and Huber
2002; Rodríguez et al. 2008), pheromones (Howard and
Blomquist 1982; Toolson 1982; Etges 1992; Greenfield 2002;
Etges et al. 2009), species-specific visual cues (Seehausen
et al. 1997; Taylor et al. 2006), host-dependent premating isolation (Dambroski et al. 2005; Feder and Nosil 2009), and
postmating isolation (Funk et al. 2002; Egan and Funk 2009;
Singer and McBride 2010), suggesting that environmentally
mediated patterns of trait expression need to be studied
when evaluating the nature of reproductive divergence and
its causes. Thus, in order to understand the evolution of
reproductive divergence and how it is expressed, phenotypic
analysis requires estimates of genetic, environmental, and G
× E interaction effects on traits thought to drive reproductive isolation and speciation, as well as how ambient ecological conditions influence patterns of gene expression.
The “top-down” approach to genetic analysis (Via 2009)
assumes that genetic influences can effectively be separated
from those of the environment, yet a central tenet of quantitative genetics is that gene expression is a consequence of
the internal and external environments in which they are
expressed. Lewontin (2000) emphasized this relationship by
asserting there is no necessary demarcation between organisms and their environments, and that organisms can define
their environments as environments can define organisms.
Thus, understanding the consequences of how and why
reproductive isolation evolves beyond characterizing the
genetic basis of traits causing isolation (Shaw and Mullen
2011) assumes an environmental context. The causal connections and feedbacks between genes, organisms, and their
environments may not be easily disentangled, and noncontextual focus on any one level, particularly DNA sequence
variation, may not resolve sources of causation (Lewontin
1991). Thus, any attempt to localize the “genes” or candidate
nucleotide substitutions (single nucleotide polymorphisms
[SNPs] or quantitative trait nucleotides [QTNs]) contributing
to speciation phenotypes is dependent on the environment(s)
in which the phenotype is assessed (Burga and Lehner 2012;
Nuzhdin et al. 2012). Consequences of this approach are, for
example, discordance between quantitative trait loci (QTL)
localization and known candidate genes (Gleason et al.
2009) and differences in QTL localization when analyses
are performed in multiple environments (Leips and Mackay
2000), as well as G × E interactions for QTL and QTNs
(Gurganus et al. 1998; Mackay and Anholt 2007; Etges et al.
2009; Stranger et al. 2011; Kang et al. 2014). Here, identifying
impacts on transcriptional variation, as well as understanding
G × E interactions in gene expression for speciation phenotypes can more efficiently help to identify sources of genetic
variation expressed in different environments (Grishkevich
and Yanai 2013). This type of functional genetic analysis can
complement genetic and genomic mapping attempts and can
be extended with proteome studies to complete genotype–
phenotype mapping.
Genotype–phenotype maps are contextual due not only
to external environments but also the internal environment,
including other genes. Genetic background effects may
reduce the generality of QTL identification and fine mapping, namely,
“Instead, the phenotype is a function of the allele, the genetic
background in which it occurs and the environment where the
mutational effects are expressed.”(Chari and Dworkin 2013)
Thus, population and species-specific differences in genetic
analyses of reproductive isolation traits are expected due to
genetic background effects, that is, epistasis (Galvan et al. 2010;
Mackay 2014), nuclear–mtDNA interactions (Pritchard et al.
2011; Burton and Barreto 2012), maternal effects, and stochastic variation (Burga and Lehner 2012) rendering general conclusions about the effects of single genes or SNPs on phenotypic
trait variation difficult unless great care is taken to assess both
population and environmental influences on expression of
reproductive isolating traits. Analysis of gene expression differences mediated by ecological variation can also be teased apart
by mapping regulatory regions or expression QTLs, but this
can be complicated by tissue-specific patterns of gene expression (Michaelson et al. 2009). Further, as natural selection operates on such standing variation, convergent evolution may result
making genotype–phenotype mapping difficult even for suspected candidate genes (Nachman et al. 2003; Rosenblum et al.
2004). Results of quantitative genetic analysis are, as always,
dependent on the environment(s) in which they are measured.
Much current interest in ecological speciation rests on inferences that causal mechanisms that drive reproductive isolation
in ecologically diverging populations can result in relatively
rapid divergence in sympatry and/or allopatry (Nosil 2012).
In the latter and more common case, allopatry with reduced
gene flow should allow for rapid divergence depending on
the strength of local adaptation. In an intensively studied species group, flies in the genus Drosophila, allopatric divergence
is thought to be the rule as few cases of sympatric speciation
have been identified (Turelli et al. 2014; Yukilevich 2014). Here,
I review progress and prospects in understanding the genetics
of premating isolating mechanisms in Drosophila mojavensis, a
cactophilic member of the genus in which ecological influences of its host plants have been shown to cause differences
in premating isolation between populations, QTL G × E interactions in traits responsible for courtship success, and patterns
of gene expression across the genome. Whole-genome microarray experiments and transcriptome sequencing have revealed
host plant effects on expression of nervous system and courtship behavior genes suggesting direct links between genome
expression and ecologically determined reproductive isolation.
Together with ongoing studies of genomic divergence, such
functional genomic data in an organism with a sequenced and
partially annotated genome should help resolve the interactions between genomes, the organism, and ecologically relevant variation in the evolution of reproductive divergence and
adaptation to different environments.
757
Journal of Heredity
Ecology of Divergence and
Reproductive Isolation in a Desert
Drosophilid
A member of the large D. repleta group, D. mojavensis is
endemic to the Sonoran Desert and adjacent arid lands in
northwestern Mexico and southern Arizona, as well as the
Mojave Desert in southern California (Heed 1978; Heed
1982; Etges et al. 1999). In Baja California, the islands in
the Gulf of California, and a small patch in coastal Sonora,
D. mojavensis uses pitaya agria cactus, Stenocereus gummosus,
almost exclusively. The remaining mainland populations in
Sonora, Sinaloa, and southern Arizona use organ pipe cactus,
S. thurberi, with occasional use of cina cactus, S. alamosensis,
with which it sometimes shares with its sibling species, D. arizonae (Fellows and Heed 1972; Markow et al. 1983; Ruiz and
Heed 1988). California barrel cactus, Ferocactus cylindraceus, is
the principal host plant in the Mojave Desert, and prickly
pears, Opuntia littoralis, O. oricola, and O. demissa (O. oricola ×
O. ficus-indica hybrids) are the only known host plants on
Santa Catalina Island near Los Angeles, California (Barbour
et al. 2007; Beckenbach et al. 2008).
The phylogeography of North American D. repleta group
species was shaped in part by the northwestward movement
of present-day Baja California on the Pacific tectonic plate
(Gastil et al. 1975) and the relatively recent formation of
the Sonoran Desert ca 13 kya (Axelrod 1979; Van Devender
1990). Along with changing sea levels forming the Gulf of
California, an effective isolating barrier between peninsular
and mainland species, D. mojavensis diverged from its closest mainland ancestor, D. arizonae, ca 1–2 mya based on Adh
and mtCOI sequence divergence (Mills et al. 1986; Matzkin
and Eanes 2003; Reed et al. 2007) and was isolated in Baja
California. Recent whole-genome comparisons of these 2
species suggest divergence was <1 mya (Lohse et al., unpublished data). This evolutionary history was first revealed
from studies of chromosomal inversions in D. mojavensis
and its closest relatives (Wasserman 1962; Ruiz et al. 1990;
Wasserman 1992) and the geographical patterns of inversion
polymorphisms throughout the range of D. mojavensis (Heed
1981; Etges et al. 1999). Ancestral populations of D. mojavensis in Baja California colonized the islands in the gulf and
mainland Sonora ca 270 kya based on coalescent analyses
of X chromosome intron sequence variation. In southern
California, Mojave Desert populations of D. mojavensis subsequently diverged from mainland populations ca 117 kya and
experienced a genetic bottleneck that reduced sequence variation 10-fold (Smith et al. 2012). The Santa Catalina Island
population represents a fourth lineage that presumably arose
from mainland California populations due to geographical
proximity, but its closest mainland relative remains unknown
(Counterman and Noor 2006; Ross and Markow 2006). Thus,
intraspecific divergence in D. mojavensis has been facilitated by
host plant shifts within the last ca 250 kya and has resulted
in host plant adaptation and partial reproductive isolation
between Baja California and mainland Mexico and Arizona
populations (Etges 1990, 1998; Etges et al. 2007, 2009, 2010).
Life-History Variation
Colonization of mainland Mexico by switching to organ
pipe cactus from agria has caused significant genetic shifts
in life history and other physiological traits in D. mojavensis
(Table 1; Etges 1990) and has resulted in premating isolation
between Baja California and mainland Mexico and Arizona
populations (Zouros and d’Entremont 1974; Wasserman
and Koepfer 1977; Zouros and d’Entremont 1980; Markow
1991). Tying adaptive shifts in life history to the evolution of
reproductive isolation suggested a role for ecological speciation in this system mediated by a host plant shift. In common
garden experiments and population crosses reared on both
host cacti, Baja California populations exhibit genetically
shorter egg to adult development times, higher viabilities, and
smaller adult body sizes than mainland Mexico and Arizona
populations (Etges 1990; Etges et al. 2010). For development
time, there were significant population × cactus and region ×
Table 1 Summary of life history and physiological differences between Baja California and mainland Mexico and Arizona populations
of Drosophila mojavensis
Phenotype
Mainland
Baja California
Region × host plant
Development time (DEVT)
DEVT homeostasis
Egg to adult viability
Thorax size
↑↑
↓↓
↓↓
↑↑
↓↓
↑↑
↑↑
↓↓
□
—
□
—
Female age at maturity
Ovariole number
Lifetime fecundity
↓↓
↑↑
↑↑
↑↑
↓↓
↓↓
□
—
—
Equivalent
↑↑
↑↑
Equivalent
↓↓
↓↓
—
—
—
Standard metabolic rate
Adult starvation resistance
EtOH vapor adult longevity
Reference
(Etges 1990; Etges et al. 2010)
(Etges 1989)
(Etges 1990; Etges et al. 2010)
(Heed and Mangan 1986;
Etges 1990; Etges 1992)
(Etges and Klassen 1989)
Heed WB, unpublished data
(Etges and Klassen 1989;
Etges and Heed 1992)
(Etges and Klassen 1989)
(Etges and Klassen 1989)
(Starmer et al. 1977; Etges
1989; Etges and Klassen 1989)
Up and down arrows show direction of differences, all statistically significant. For experiments based on cactus-cultured populations, presence of significant
region × host plant interactions is also shown.
758
Etges • Genetics, Environments, and Speciation
cactus interactions, and for egg to adult viability, a significant
region × cactus interaction revealed host plant adaptation
where Baja California populations expressed higher viability
on agria versus organ pipe cactus while mainland populations expressed higher viability on organ pipe cactus, host
plants these populations use in nature (Table 1). In many replicate population crosses, F1, F2, and backcross generations
showed higher egg to adult viability than parental populations particularly when reared on agria cactus revealing hostspecific heterosis for fitness. Host plant, additive, dominance,
and cytoplasmic effects were detected for development time,
as well as a significant X chromosome effect associated with
the typically shorter development times of Baja California
populations. Higher egg to adult viability in Baja flies showed
a significant X chromosome effect, but this was not observed
in both sets of populations crosses suggesting geographic
variation in X-linked genes that increase fitness relative to
mainland populations (Etges et al. 2010).
Reproductive Isolation
Premating and postmating-prezygotic isolation (Knowles
and Markow 2001) between Baja California and mainland Mexico and Arizona populations have long suggested
that allopatric populations of D. mojavensis are currently in
the initial stages of reproductive isolation making them an
ideal system with which to explore the early stages of divergence where no postmating isolation has arisen (Ruiz et al.
1990). A causal relationship between host plant adaptation
and reproductive isolation predicted from the hypothesis
of ecological speciation suggested a testable alternative to
earlier inferences about premating isolation between the
Baja mainland populations of D. mojavensis. Previous studies suggested the cause was due to reproductive character
displacement because of the presence of D. arizonae on the
mainland, but not in Baja California, causing shifts in mate
choice of mainland D. mojavensis (Mettler and Nagle 1966;
Wasserman and Koepfer 1977). The latter hypothesis has
not been tested experimentally in part because these species
hybridize in the laboratory (Mettler 1957; Jennings and Etges
2010) but do not currently hybridize in nature (Etges et al.
1999; Counterman and Noor 2006; Machado et al. 2007), so
reinforcement is unlikely to be involved. Laboratory multiple-choice tests involving Mojave Desert and Santa Catalina
Island populations with Baja California and mainland Mexico
populations revealed little evidence for sexual isolation
between California populations with either Baja California
or mainland populations with few exceptions (Markow 1991;
Massie and Markow 2005). Also, D. arizonae populations sympatric with D. mojavensis do not exhibit increased premating
isolation with other populations, so there may have been the
evolution of local mate recognition systems unrelated to the
presence of D. arizonae. A more parsimonious explanation,
historical gene flow between species, was suggested from
emerging patterns of genomic divergence between allopatric
and sympatric populations of D. arizonae and D. mojavensis,
suggesting that interspecific hybridization ceased ca 200–250
kya (Lohse et al., unpublished data). Why these genomic
differences were not detected in earlier studies (Counterman
and Noor 2006; Machado et al. 2007) maybe due to a lack of
power due to smaller numbers of loci studied or the choice
of divergence with gene flow models that were used (Lohse,
personal communication).
Observations that premating isolation between Baja
California and mainland populations is significantly influenced by rearing substrates (Brazner 1983; Etges 1992) suggested the preadult rearing environments determined the
strength of assortative mating between populations, and
hence is critical to understanding divergence and reproductive isolation. Rearing D. mojavensis on laboratory media significantly increases the strength of sexual isolation causing
decreased mating propensity of Baja California males and a
dearth of matings between mainland females and Baja males
in multiple-choice tests. Cactus-reared flies exhibit lower levels of sexual isolation, with agria-reared flies typically showing little or no premating isolation and organ pipe-reared flies
showing low, but significant levels of behavioral isolation
(Brazner and Etges 1993). Further, sexual isolation between
D. mojavensis and D. arizonae increased when flies were reared
on cactus versus those reared on laboratory food and when
exposed to fermenting host plant tissue during mating trials (Jennings and Etges 2010). Thus, for these sibling species as well as allopatric, ecologically diverged populations of
D. mojavensis, understanding the earliest stages of reproductive isolation requires knowledge of the consequences of
host plant use.
As in many drosophilids, courtship decisions result from a
progressive, multistage interaction between males and females
involving visual, auditory, and tactile cues (Greenspan and
Ferveur 2000). In D. mojavensis, courtship success is determined mainly by female choice (Havens et al. 2011) based on
variation in male courtship songs (Byrne 1999; Etges et al.
2006) and cuticular hydrocarbons (Markow and Toolson
1990; Etges and Ahrens 2001). Selection gradient analysis
revealed that variation in multiple components of courtship
songs, that is, long interpulse interval (L-IPI), burst duration,
interburst interval (IBI), and number of bursts produced all
influenced male courtship success with mainland females,
but rearing substrates had no additional effects on any component of courtship song on mating success (see Table 1 in
Etges et al. 2007). Mainland females preferred copulating
with males that produce songs with shorter L-IPIs, shorter
burst durations, and shorter IBIs. However, courtship success was significantly higher in males reared on organ pipe
cactus, was influenced by a single detected QTL very near
the desat-2 locus, and 4 G × E interactions (Etges et al. 2007).
In a common garden experiment with 5 mainland populations and 6 Baja populations reared on agria and organ pipe
cactus, few cuticular hydrocarbons (CHCs) showed effects
of rearing substrate effects, but differences due to sex and
region were strongly significant including a number of sex ×
region interactions for CHCs associated with mating success.
These results suggested that male courtship songs and contact pheromones are part of Baja California versus mainland
mate recognition systems in D. mojavensis (Etges and Ahrens
2001).
759
Journal of Heredity
Allopatry and reduced gene flow between diverging
populations should allow for rapid, evolution of local mate
recognition, that is, sexual selection, postmating-prezygotic
mechanisms, and drivers of sexual conflict (Williams 1966).
As gene flow between Baja California, mainland Mexico, and
southern California populations is insignificant or too low to
be detected (Smith et al. 2012), genomic divergence should
be unconstrained by possible effects of immigrant alleles that
are poorly adapted locally (Nosil et al. 2005). So, genomic
regions containing loci associated with ecological adaptation
are not expected to be closely linked with genes inflecting
reproductive isolation or be located within inversions protecting them from recombination (Feder et al. 2012; Nosil
et al. 2012). Thus, divergence genetics of premating isolation
should reveal how the initial stages of speciation evolve in
allopatric populations of D. mojavensis.
Both male courtship songs and CHCs are polygenic traits
(Gleason et al. 2002, 2005; Howard and Blomquist 2005), so
an understanding of their genetic architectures, as well as the
genetic basis of life-history variation, is necessary to begin to
understand the forces that have shaped their variation (Shaw
and Parsons 2002). In a test of the ecological speciation
hypothesis, a genetic correlation was observed between variation in egg to adult development time and premating isolation (Etges 1998), suggesting an alternate explanation for the
evolution of premating isolation in D. mojavensis and that even
in allopatry, regions of the genomes associated with ecological adaptation can overlap those associated with reproductive isolation (cf., Hawthorne and Via 2001). We still do not
know the nature of this association in D. mojavensis, that is,
how many loci are involved or whether pleiotropy or linkage
is responsible, but genetic evidence derives from 4 separate
experimental observations. First, this genetic correlation was
identified in a bidirectional artificial selection experiment on
egg to adult development time in Baja California and mainland populations of D. mojavensis reared on agria and organ
pipe cactus. Significant selection responses in development
time with realized heritabilities of 8–11% were observed
together with correlated responses in decreased premating isolation between populations (Etges 1998). Second, in
populations cultured in a common garden experiment to
assess variation in CHCs across the species range (Etges and
Ahrens 2001), adults from mainland populations with longer
development times had significantly more CHCs than adults
that eclosed earlier (Figure 1). Third, F2 males that eclosed
on the first 2 days from cactus cultures and aged to sexual
maturity had less CHCs than those eclosing later, although
this was influenced by exposure/no exposure to females
(Etges et al. 2009). Four, QTL influenced both CHCs and
development time from the same large-scale genetic analysis
of cactus-reared F2 males. QTL genotype scores for development time and CHC variation known to influence mating
Figure 1. Replicate CHC samples from adults pooled by sex and rearing substrates of mainland populations from Punta Onah
and Rancho Diamante, Sonora. All adults in each replicate culture that emerged on the first 2 days were labeled “fast” and the
remaining flies labeled “slow.” Amounts of 11 CHC components were significantly greater in the “slow” groups (n = 24) than
in the “fast” groups (n = 22) using paired t-tests (all P ≤ 0.02). Total hydrocarbons per fly were also greater in the “slow” groups
than in the “fast” groups (3075.5 vs. 2632.3 ng/fly, t = 2.38, P < 0.022). CHC components are labeled by equivalent carbon chain
lengths defined in Etges and Ahrens (2001).
760
Etges • Genetics, Environments, and Speciation
success between Baja and mainland D. mojavensis were positively correlated, r = 0.59, P = 0.02 (Etges et al. 2010). Two
of these QTL, marked by microsatellite loci Dmoj2010 and
Dmoj2_6540c, are located on the second chromosome with
the latter very near fruitless, a gene with known effects on
mating behavior (Ryner et al. 1996; Greenspan and Ferveur
2000). Dmoj2010 is downstream from crossveinless c that functions in lipid binding and multiple developmental processes.
The other 2 are on different chromosomes, Dmoj3030 that
is very close to GI10093: its Drosophila melanogaster ortholog
encodes a type of immunoglobulin, and Dmoj4300 is near
GI13245, an unannotated protein coding gene. Further fine
scale mapping of these QTL regions, given the mostly annotated genome of D. mojavensis (Tweedie et al. 2009), is of
high priority to assess the causes for this genetic correlation
between development time, premating isolation, and CHCs.
Host Plant Effects on Gene Expression
A clearer genetic understanding of genetic and environmental influences on phenotypic variation is now possible
using whole genome expression patterns at either the transcriptome or proteome levels (Gibson 2008; Pavey et al.
2012; Wray 2013). Here, we tested the hypothesis that host
cacti cause differences in expression of genes known to be
involved with courtship success as suggested by the previous observations of differences in premating isolation and
sexual selection caused by rearing populations of D. mojavensis on agria or organ pipe cactus. We investigated the effects
of different host plants on patterns of whole transcriptome
expression to directly identify genes involved with courtship
success, songs, and CHCs. These approaches offer the exciting opportunity to answer the ultimate questions in evolutionary genetics concerning the differential expression of genes
and genomic elements responsible for divergence in speciation phenotypes. In nonmodel systems, expressed sequence
tags can be used (Frentiu et al. 2009; Schwarz et al. 2009),
but identification of genetic elements of known function,
that is, annotated genes, is critical to inferences about genetic
architecture allowing understanding of the consequences of
ecological divergence on reproductive isolation. Microarray
technology (Quackenbush 2001) and RNAseq methods
(Oshlack et al. 2010) both have advantages/disadvantages,
but functional genomic inferences from both techniques are
limited by the extent of genome annotation for most organisms. Microarray design allows for highly controlled hybridization with each oligonucleotide on a predesigned array and
does not require population samples to be inbred. RNAseq
is not limited to coding regions and can identify patterns of
alternate exon splicing (Malone and Oliver 2011), but is sensitive to any residual sample heterozygosity, that is, inbreeding is required.
Host plant effects on expression of life-history variation
in D. mojavensis (Etges 1990; Etges et al. 2010), premating
isolation between Baja California and mainland populations
(Etges 1992; Brazner and Etges 1993), male mating success
mediated by courtship songs (Etges et al. 2007), and QTL
analysis of CHCs (Etges et al. 2009) all suggest that revealing
host plant mediated differences in gene expression should
help to identify how rearing substrates have such large effects
on life-history traits and courtship behavior. Here, data from
2 experiments demonstrate the efficacy of this approach for
1) adult virgin females exposed to fermenting cactus throughout adulthood (Etges et al., submitted), and 2) cactus-reared
males that were either successful or unsuccessful in copulating with females in mating choice trials (for details, see Smith
et al. 2013).
For the first experiment, we used gene predictions
from the D. mojavensis reference genome (Drosophila-12Genomes-Consortium 2007) to design a 12-plex microarray
from Roche-Nimblegen containing 14 528 unique oligomers
corresponding to different genes. We designed replicated laboratory experiments to investigate effects of cactus-rearing
and population differentiation on life cycle stage/age differences in gene expression in order to identify candidate
genes that were differentially expressed over the life cycle.
Array design, hybridization protocols, and data analysis are
described in Rajpurohit et al. (2013), but only 9114 coding
regions (63%) are orthologous with D. melanogaster genes
limiting our interpretations of differential gene expression
and gene ontology in D. mojavensis. Far fewer genes, ca 75,
have been directly annotated in D. mojavensis. Nevertheless,
the results of these experiments have revealed the identity
of genetic systems with inferred biological functions that
differed in gene expression in response to those factors
suspected to have influenced the diversification of geographically differentiated populations that use different host cacti
(Etges et al., submitted).
A mainland D. mojavensis population from Las Bocas,
Sonora, and a Baja California population from Punta Prieta
were cultured on both fermenting agria and organ pipe cactus from eggs to 28-day old adults such that flies were never
exposed to lab food. All preadult life stages and adults of different ages adults were sampled for analysis of gene expression using whole genome microarrays. Groups of 30 adult
flies that had been reared on fermenting agria or organ pipe
cactus (Etges 1992) were transferred to 25 × 95 mm glass
tubes on the day of eclosion. The ends of each tube were
plugged with fine mesh to allow air circulation at one end
and a feeding cup containing either fermenting agria or organ
pipe cactus at the other. Every day, the tubes were placed in
sealed desiccators containing 1 L of 4% ethanol from 8:00
AM to 6:00 PM allowing adults to feed on ethanol vapor. For
the remaining 14 h each day, all tubes were removed from
each desiccator and kept in the incubator set to a 14:10 LD
photoperiod and 27:17 °C to minimize condensation inside
the tubes. Every 4 days, the food cups were replaced.
At 0, 3, 6, 10, 14, 18, 24, and 28 days of age, groups of 24
adult females were snap frozen in liquid nitrogen resulting in
an experimental design of 2 populations × 2 host cacti × 7
ages (24- and 28-day old flies were pooled due to small numbers of surviving 28-day adults) × 4 replicates for RNA processing and microarray hybridization. General experimental
details, bioinformatic approaches, and statistical approaches
are detailed in Rajpurohit et al. (2013). In short, we assessed
761
762
False Discovery rate P values, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.
in parentheses indicate the percentage of genes with annotation in Drosophila melanogaster.
aNumbers
Significantly enriched GOTerms (P < 0.05) are indicated. Each population comparison denotes genes that were overexpressed in each population. Mainland population LB = Las Bocas, Baja California population
PP = Punta Prieta. There were no cases of significantly greater gene expression for the agria > organ pipe cactus comparison. FDR, false discovery rate.
1.5
2. Circadian
rhythm, courtship
behavior
2.2*
2.1****
2. Iron/
heme binding,
oxidation
reduction
2. Fatty acid
biosynthesis
2.3***
1. Iron/heme
binding/P450
1. Transcription/
RNA
biosynthetic
process
2.7****
1. Neurotransmitter
binding/receptors
2.2**
1. Fatty acid/
lipid biosynthesis
2.5***
Enrich score
GOTerm
Enrich score
Enrich score
GOTerm
GOTerm
Enrich score
GOTerm
Number of genes = 514,
number annotated = 302 (62.3)
Number of genes = 66,
number annotated = 45 (68.2)
Number of genes = 251,
number annotated = 146 (58.2)
Number of genes = 184,
number annotated = 79 (42.9)a
Cactus × population
Cactus (organ pipe > agria)
Population (PP > LB)
Population (LB > PP)
age-specific differences in gene expression using false discovery rate corrected paired t-tests (Benjamini and Hochberg
1995) that had absolute fold changes >1.5. Using the 9114
orthologs from D. melanogaster, we performed gene ontology
analysis with DAVID Bioinformatics Resources 6.7 (Huang
et al. 2009) in order to identify functionally related genes that
changed with age. For the present analysis focusing on cactus effects, we pooled the data across adult ages and performed a mixed model Anova in SAS (SAS-Institute 2004)
with population, cactus, and population × cactus effects on
gene expression (Table 2, Supplementary Table 1). We limited our analyses to adult females due to the large number of
whole genome hybridizations (n = 86, some replicates were
missing).
Organ pipe-reared adults showed significantly increased
expression of genes that were enriched for neurotransmitter
binding, circadian rhythm, and courtship and mating behavioral functions, but no genes were expressed at significantly
greater levels due to agria than organ pipe cactus (AG > OP)
for females from these 2 populations (Tables 2 and 3). Thus,
organ pipe cactus increased expression of a small group of
gated ion channel genes associated with neurotransmission,
circadian rhythm, and courtship behavior genes in female
D. mojavensis. One odorant binding gene, Obp56h, was also
upregulated due to organ pipe cactus, showed significant
increases in transcription starting in the first larval instar
(Etges et al., submitted), is expressed in antennae of both
sexes, the pharyngeal organs, and possibly functions in both
olfactory and gustatory systems (Galindo and Smith 2001),
but ligands that it interacts with remain unknown. Given that
organ pipe-reared females, particularly those from mainland
populations, express increased discrimination among males
during courtship (Etges 1992; Brazner and Etges 1993), and
organ pipe-reared F2 males had higher mating success than
those reared on agria cactus and exhibited G × E interactions
for CHCs (Etges et al. 2009), these groups of genes (Table 3)
are good candidates for further analysis of why organ pipe
cactus increases premating isolation between Baja and mainland populations of D. mojavensis.
These results also indicated that mainland, Las Bocas
females showed significant enrichment for fatty acid metabolism genes, as well as genes with iron binding functions, such
as P450 genes associated with detoxification of xenobiotics,
secondary compounds in these cacti including fatty acids,
sterol diols, and high levels of triterpene glycosides (Fogleman
et al. 1998). Of the 251 genes showing higher expression in
Punta Prieta versus Las Bocas females, 146 (58.2 %) were
annotated and showed significant enrichment for transcription/RNA synthesis indicating these Baja females had higher
overall rates of gene expression than mainland females when
reared on fermenting cactus and ethanol vapor (Table 2). For
transcript abundance showing significant cactus × population interactions, the 302/514 annotated orthologs were significantly enriched for iron binding/P450 function and fatty
acid synthesis similar to the LB > PP population level differences (Table 2). Overall, the patterns of differential gene
expression in these populations of D. mojavensis revealed that
differences in cactus rearing substrates had significant effects
Table 2 Significant overall adult gene expression differences (FDR <0.01 and >1.5-fold difference) due to population, cactus, and population × cactus interaction
Journal of Heredity
Etges • Genetics, Environments, and Speciation
Table 3 Orthologous Drosophila melanogaster genes that were significantly overexpressed in adult Drosophila mojavensis reared on organ
pipe versus agria cactus that comprised the 2 significant enriched GO clusters from Table 2, 1) neurotransmitter binding/receptors and
2) circadian rhythm, courtship behavior
1.
2.
Dmel gene ID
Dmel orthologue
Dmel ortho name
Biological function
FBgn0034656
CG17922
FBgn0004244
Rdl
Intracellular cyclic nucleotide
activated cation channel
complex
Resistant to dieldrin
FBgn0032151
nAcRα-30D
FBgn0024963
GluClα
FBgn0028875
nAcRα-34E
Nicotinic acetylcholine
receptor α 34E
FBgn0004244
Rdl
Resistant to dieldrin
FBgn0027491
Cdk5α
Cdk5 activator-like protein
FBgn0031652
FBgn0039298
jet
to
jetlag
takeout
FBgn0015374
crl
Courtless
FBgn0004595
pros
prospero
FBgn0034475
Obp56h
Odorant-binding protein 56h
Nicotinic acetylcholine
receptor α 30D
GluClα
Regulation of olfactory learning, regulation of
circadian sleep/wake cycle
Ion transport
Extracellular-glutamate-gated chloride channel
activity, postsynaptic membrane
Acetylcholine binding, acetylcholine-gated
channel complex
Regulation of olfactory learning, regulation of
circadian sleep/wake cycle
Locomotor behavior, axonogenesis,
determination of adult lifespan
Circadian behavior, locomotor rhythm
Courtship behavior, adult feeding behavior,
circadian rhythm
Ubiquitin-protein ligase activity, male courtship
behavior, spermatogenesis
Sequence-specific DNA binding transcription
factor activity, nervous system development
Olfactory behavior, response to pheromone
See text for details.
on nervous system and courtship behavior genes and that
these populations responded differently to these host plants
in transcription of genes involved with xenobiotic metabolism and fatty acid synthesis.
In the second study, Smith et al. (2013) used SOLiD 4
transcriptome sequencing to assess agria and organ pipe
cactus-reared males allowed to court and/or mate with
organ pipe-reared females in multiple-choice tests. All
flies were derived from a population collected in Organ
Pipe National Monument, Arizona. Differences due to
cactus-rearing substrates, male mating success and mating success × cactus interactions were assessed for both
differentially expressed transcripts and those showing
alternative splicing. However, these males were cultured
on fermenting cactus, but reared on lab food as adults
until the mating trials began as in previous studies (Etges
1992). Just 19 orthologs showed differences in expression levels due to mating success, and only 11 of these
were annotated (Tables 4 and 5). In males that achieved
copulation, Dmoj\GI18531, an ortholog of CG16894
in D. melanogaster, was upregulated that is associated with
ubiquitin-conjugating enzyme activity in testis. Another
protein-binding gene, an ortholog of Traf4, was upregulated, as were grg (no known function), esg associated with
olfactory behavior regulation of transcription though zinc
finger DNA binding domains, and CG1887 associated
with defense response. Accessory protein gene Acp32CD
was upregulated in unmated males; its function is associated with decreasing female receptivity suggesting male
D. mojavensis decrease transcription of Acp32CD after
mating.
For rearing substrate differences, I reanalyzed the data in
Smith et al. (2013) by separating the gene lists for differential
expression to assess relative up/down transcript abundance
(Tables 4 and 5) as they grouped these genes for functional annotation. As with cactus-reared females described above, far more
genes showed significantly greater transcript abundance in organ
pipe-reared males versus agria-reared males, OP > AG, n = 97,
than AG > OP, n = 14. Although Smith et al. (2013) reported
4 genes enriched for chemosensory and olfactory function, this
functional group was not recovered when the up/down lists
were separated (Tables 4 and 5). Only 44% of these genes are
annotated and showed weak functional clustering for nucleotide
binding and signal transduction. Agria cactus caused increased
transcript abundance for 14 genes, 11 of which were annotated,
with most involved in defense and immune response (Smith
et al. 2013). A larger number of differentially expressed genes
exhibited cactus × mating success interactions, n = 147, that were
weakly enriched for protein manufacture including mitochondrial
function and assembly, as well as ribonuclear binding.
Transcriptome sequencing made possible analysis of differences in alternate exon splicing (AE) influenced by mating
success and cactus rearing (Table 5). An unexpectedly large
number of genes with AE differences were observed caused
by rearing substrates (n = 64) and the interaction between
cactus and mating success (n = 48) that were functionally
enriched for membrane ion channel, metal ion binding, and
nucleotide binding. Gene annotations revealed multiple cases
763
764
aResults
SerB
phosphatase
Regulation
female
receptivity
RNA pol II
Phosphatase
Antibacteria
resp.
GTPase
.
Olfactory
behavior
G-protein
coupled receptor
0.9a
0.6
0.7
from Smith et al. (2013), for cactus they did not separate up/down gene lists.
Rab30
Bgb
CG11425
DptB
Acp32CD
aay
Traf4
Mated < unmated = 10
grg
esg
Defense
response
No known function
Olfactory
behavior
Protein binding
CG1887
Nucleotide
binding
GOTerm
Mated > unmated = 9
Ubiquitin
enzyme
Number annotated = 43 (44.3%)
Number annotated = 11 (57.9%)
CG16894
Number of genes = 97
Number of genes = 19
Enrich score
Cactus (organ pipe > agria)
Mating success
Table 4 Differences in transcript abundance
No functionally
enriched clusters,
most genes are
defense response
GOTerm
Enrich score
Number annotated = 11 (78.6%)
Number of genes = 14
Cactus (agria > organ pipe)
Nucleotide
binding, RNA
recognition
Ribosome
Mitochondria
GOTerm
0.7
1.1
1.1
Enrich score
Number annotated = 82 (62.3%)
Number of genes = 147
Cactus × mating success
Journal of Heredity
Etges • Genetics, Environments, and Speciation
Table 5 Alternate splicing in adult Drosophila mojavensis males that either successfully mated or did not
Mating success
Cactus
Cactus × mate
Number of genes = 28
Number of genes = 64
Number of genes = 48
Number annotated = 23 (82.1)
Number annotated = 49 (76.5)
Number annotated = 33 (68.8)
GOTerm
GOTerm
Enrich score
GOTerm
Enrich score
Ion channel,
membrane
Ion, metal binding
1.4
Ion, metal binding
0.4
1.0
Nucleotide binding
0.2
Enrich score
None
Gene names are Drosophila melanogaster orthologs and GOTerms are functionally enriched clusters identified by DAVID. See text for details.
Table 6 Functional clustering of alternately spliced genes due to the effects of culturing male Drosophila mojavensis on organ pipe versus
agria cactus
Dmoj gene
Dmel ortho
Dmel ortho name
Biological function
GOTerm–ion channel, membrane
FBgn0132826
CG6723
FBgn0140123
Ca-alpha1D
—
Ca-alpha1D
FBgn0142139
CG7882
—
FBgn0146694
nAcRalpha-80B
FBgn0146742
slo
Nicotinic acetylcholine
receptor α 80D
Slowpoke
FBgn0147216
Nmdar1
Sodium iodide symporter activity
Voltage-gated calcium channel
complex
Glucose transmembrane
transporter activity
Nicotinic acetylcholine-gated
receptor channel complex
Calcium-activated potassium
channel activity, courtship songs
N-methyl-D-aspartate selective
glutamate receptor activity, longterm memory
GOTerm–ion, metal binding
FBgn0133359
CG17272
FBgn0137845
rdgB
Glutamate NMDA
receptor subunit 1
—
Retinal degeneration B
FBgn0138983
peb
Pebbled
FBgn0142664
ari-2
Ariadne 2
FBgn0142910
Git
FBgn0146608
Rim
FBgn0146742
slo
G protein-coupled
receptor kinase
interacting ArfGAP
Rab3-interacting
molecule
Slowpoke
FBgn0146746
FBgn0147079
FBgn0147489
tld
CG14869
mRpL19
FBgn0147216
Nmdar1
Tolloid
—
Mitochondrial
ribosomal protein L19
Glutamate NMDA
receptor subunit 1
of cactus-caused AE for ion membrane channel function
associated with memory and courtship song (Table 6), including slowpoke, known to influence pulse rate in D. melanogaster
courtship songs (Peixoto and Hall 1998) due to its role in
calcium-activated potassium channel activity. While none of
these cactus-influenced AE genes overlapped with those that
were differentially regulated in females (Table 3), there were
Calcium ion binding
Rhodopsin-mediated signaling
pathway, phototransduction,
perception of smell
Transcription factor activity, zinc
finger, C2H2-like
Ubiquitin protein ligase activity,
zinc finger, C6HC type
Somatic muscle development, ARF
GTPase activator activity
Small GTPase regulator activity,
regulation of exocytosis
Calcium-activated potassium
channel activity, courtship songs
Metalloendopeptidase activity
Metalloendopeptidase activity
Mitochondrial large ribosomal
subunit
N-methyl-D-aspartate selective
glutamate receptor activity, longterm memory
a few like nAcRα-34E and nAcRalpha-80B, both with acetylcholine-gated receptor function, and Nmdar1 and GluClα,
both with glutamate receptor activity and gated chloride channel activity, that exhibited functional overlap associated with
neurotransmitter binding/receptor activity. Thus, differences
in gene expression due to cactus-rearing substrates influenced
males and females in somewhat similar ways involving neural
765
Journal of Heredity
function and gated ion channel activity including variation in
expression of calcium-activated potassium channels controlled
by the slowpoke protein thought to alter flight muscle excitability and several courtship song components (Greenspan and
Ferveur 2000).
In D. mojavensis, a handful of QTL and QTL × cactus interactions influenced courtship song variation. The second chromosome region containing slowpoke, linked to microsatellite
locus Dmoj2_2868a, was associated with variation in male time
to copulation; P = 0.01, components of courtship song variation including interburst interval, IBI; P = 0.003, and number
of bursts; P = 0.009 (Etges et al. 2007). IBI and number of
bursts were also influenced by Dmoj2_2868a × cactus interactions (P = 0.036, and 0.039 respectively). Thus, in cactusreared D. mojavensis, male courtship song differences expressed
in a cactus-specific manner were associated with both genetic
differences between mainland and Baja populations and were
localized around slowpoke, as well as cactus influenced differences in patterns of alternate messenger RNA splicing involving slowpoke and other genes. Without the environmental
context of host cactus differences, it would not be possible to
both investigate why cactus-specific sets of genes are differentially expressed in males and females, and certainly be more
difficult to begin isolating candidate genes like slowpoke in order
to investigate how alternate splicing patterns caused by different host plants may contribute to differences in courtship success and premating isolation between populations.
Prospects for Speciation Genetics
A major goal in most speciation research programs is to isolate the genetic factors responsible for reproductive isolation.
In systems where ecological divergence can be shown to be
associated with adaptations that have caused reproductive
isolation thought to be driving speciation, an ultimate goal is
to identify and understand the particular sets of interacting
genes and their patterns of expression that have evolved in
response to differences in both sexual and ambient environmental conditions. Whole-genome transcriptomic analysis is
a first step in understanding the nature of genome–organism–environment interactions for speciation phenotypes, but
much more work will be necessary to understand the subtle
connections between gene expression and phenotypic variation. At the very least, we have isolated groups of differentially and alternately expressed candidate genes that are
sensitive to rearing on alternate host plants. As these host
plant effects are known to directly influence the degree of
premating isolation mediated by CHCs and courtship songs,
this approach has brought us closer to identifying functional
groups of genes responsible for reproductive isolation.
Given the long-term interest in D. mojavensis as a model species for understanding interactions between its patterns of
host plant use (Fellows and Heed 1972), patterns of premating isolation (Zouros and d’Entremont 1974; Wasserman and
Koepfer 1977), phylogeography (Heed 1978), history (Ruiz
et al. 1990), use of cactophilic yeasts and bacteria (Starmer
et al. 1976; Young et al. 1981; Fogleman and Foster 1989),
766
adaptation to host plants (Etges 1990), and systematic relationships (Durando et al. 2000; Oliveira et al. 2012), these
efforts in understanding genome evolution and expression
in different environments will reveal the genetic basis for
adaptation and reproductive isolation, and how they are connected. Both regulatory and structural genetic elements are
expected to be involved, so faster progress will be made when
fully sequenced and assembled genomes are made available
that have been functionally annotated. Although the logic of
gene ontology relies on critical assumptions about structure
and function that may not always hold as structurally similar
genes may evolve new functions in other species, it remains a
major tool in functional genomics.
Because functional gene clusters have been identified that
influence expression of traits responsible for reproductive
isolation in the context of the environments in which they
are expressed, we expect further progress will be made by
isolating downstream expression differences, proteomics,
and then assembling protein networks in the context the of
biochemical pathways and tissue-specific patterns of expression. As most of the phenotypes of interest are expected
to be polygenic, a gene network approach will be required
as classical genetic tools in nonmodel species such as gene
knockouts are not yet available. Identification of candidate
genes with known physiological functions can, however,
speed genetic analysis to reveal examples of simpler genotype–phenotype relationships with potential relevance to
ecological divergence and reproductive isolation (e.g., Chung
et al. 2014).
Perhaps the greatest impediment to revealing underlying
mechanisms in speciation genetics in almost all organisms is
the lack of fully annotated genomes. A major challenge to
complete the detailing of the interactions between genomes
and the environments in which they are expressed is to understand the functional interaction of all genetic elements that
contribute to reproductive isolation. Without documented
gene functions in the organisms in which they are expressed,
this goal will remain elusive.
Supplementary Material
Supplementary material can be found at http://www.jhered.
oxfordjournals.org/.
Funding
National Science Foundation grants (NSF DEB-02111025 to
W.J.E., EF-0723930 to A. G. Gibbs and W.J.E.). Microarray
processing at the UNLV Genomics Core Facility was supported by grants from the National Center for Research
Resources (5P20RR016464-11); the National Institute of
General Medical Sciences (8P20GM103440-11).
Acknowledgments
I want to thank K. Shaw for organizing the 2013 AGA symposium on
speciation and making this collection of symposium papers possible, A. J.
Etges • Genetics, Environments, and Speciation
Marlon and G. Almeida for excellent technical assistance, S. Rajpurohit and
A. Gibbs for microarray data processing and discussion, and C. Ross for
statistical help with the microarray studies. MIAME compliant microarray
data is deposited at Gene Expression Omnibus (http://www.ncbi.nlm.nih.
gov/geo/) with accession number GSE54510.
References
Axelrod DI. 1979. Age and origin of Sonoran Desert vegetation. Calif Acad
Sci Occas Papers. 132:1–74.
Barbour M, Keeler-Wolf T, Schoenherr AA. 2007. Terrestrial vegetation of
California. Berkeley (CA): University of California Press.
Beckenbach AT, Heed WB, Etges WJ. 2008. A mitochondrial DNA analysis of vicariant speciation in two lineages in the Drosophila mulleri subgroup.
Evol Ecol Res. 10:475–492.
Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: a
practical and powerful approach to multiple testing. J Royal Stat Soc B.
57:289–300.
Drosophila-12-Genomes-Consortium. 2007. Evolution of genes and
genomes on the Drosophila phylogeny. Nature. 450:203–218.
Durando CM, Baker RH, Etges WJ, Heed WB, Wasserman M, DeSalle
R. 2000. Phylogenetic analysis of the repleta species group of the genus
Drosophila using multiple sources of characters. Mol Phylogenet Evol.
16:296–307.
Egan SP, Funk DJ. 2009. Ecologically dependent postmating isolation
between sympatric host forms of Neochlamisus bebbianae leaf beetles. Proc
Natl Acad Sci USA. 106:19426–19431.
Etges WJ. 1989. Divergence in cactophilic Drosophila: the evolutionary significance of adult ethanol metabolism. Evolution. 43:1316–1319.
Etges WJ. 1989. Evolution of developmental homeostasis in Drosophila
mojavensis. Evol Ecol. 3:189–201.
Etges WJ. 1990. Direction of life history evolution in Drosophila mojavensis.
In: Barker JSF, Starmer WT, MacIntyre RJ, editors. Ecological and evolutionary genetics of Drosophila. New York: Plenum. p. 37–56.
Etges WJ. 1992. Premating isolation is determined by larval rearing substrates in cactophilic Drosophila mojavensis. Evolution. 46:1945–1950.
Blair WF. 1947. Variation in shade of pelage of local populations of the
cactus-mouse (Peromyscus eremicus) in the Tularosa basin and adjacent areas of
southern New Mexico. Contr Lab Vert Biol Univ Michigan. 37:1–7.
Etges WJ. 1998. Premating isolation is determined by larval rearing substrates
in cactophilic Drosophila mojavensis. IV. Correlated responses in behavioral
isolation to artificial selection on a life history trait. Am Nat. 152:129–144.
Bradshaw HD, Wilbert SM, Otto KG, Schemske DW. 1995. Genetic mapping of floral traits associated with reproductive isolation in monkeyflowers
(Mimulus). Nature. 376:762–765.
Etges WJ, Ahrens MA. 2001. Premating isolation is determined by larval
rearing substrates in cactophilic Drosophila mojavensis. V. Deep geographic
variation in epicuticular hydrocarbons among isolated populations. Am Nat.
158:585–598.
Brazner JC. 1983. The influence of rearing environment on sexual isolation
between populations of Drosophila mojavensis: an alternative to the character
displacement hypothesis [MS]. Syracuse: Syracuse University.
Brazner JC, Etges WJ. 1993. Pre-mating isolation is determined by larval
rearing substrates in cactophilic Drosophila mojavensis. II. Effects of larval
substrates on time to copulation, mate choice, and mating propensity. Evol
Ecol. 7:605–624.
Burga A, Lehner B. 2012. Beyond genotype to phenotype: why the
phenotype of an individual cannot always be predicted from their
genome sequence and the environment that they experience. FEBS J.
279:3765–3775.
Burton RS, Barreto FS. 2012. A disproportionate role for mtDNA in
Dobzhansky–Muller incompatibilities? Mol Ecol. 21:4942–4957.
Byrne BC. 1999. Behaviour-genetic analysis of lovesongs in desert species
of Drosophila [Ph.D.]. Leicester: University of Leicester.
Chari S, Dworkin I. 2013. The conditional nature of genetic interactions:
the consequences of wild-type backgrounds on mutational interactions in a
genome-wide modifier screen. PLoS Genet. 9:e1003661.
Chung H, Loehlin DW, Dufour HD, Vaccarro K, Millar JG, Carroll SB.
2014. A single gene affects both ecological divergence and mate choice in
Drosophila. Science. 343:1148–1151.
Cook LM, Grant BS, Saccheri IJ, Mallet J. 2012. Selective bird predation on the
peppered moth: the last experiment of Michael Majerus. Biol Lett. 8:609–612.
Counterman BA, Noor MA. 2006. Multilocus test for introgression between
the cactophilic species Drosophila mojavensis and Drosophila arizonae. Am Nat.
168:682–696.
Etges WJ, de Oliveira CC, Gragg E, Ortíz-Barrientos D, Noor MA, Ritchie
MG. 2007. Genetics of incipient speciation in Drosophila mojavensis. I. Male
courtship song, mating success, and genotype x environment interactions.
Evolution. 61:1106–1119.
Etges WJ, de Oliveira CC, Noor MA, Ritchie MG. 2010. Genetics of incipient speciation in Drosophila mojavensis. III. Life-history divergence in allopatry
and reproductive isolation. Evolution. 64:3549–3569.
Etges WJ, de Oliveira CC, Ritchie MG, Noor MAF. 2009. Genetics of
incipient speciation in Drosophila mojavensis. II. Host plants and mating status
influence cuticular hydrocarbon QTL expression and G x E interactions.
Evolution. 63:1712–1730.
Etges WJ, Heed WB. 1992. Remating effects on the genetic structure of
female life histories in populations of Drosophila mojavensis. Heredity.
68:515–528.
Etges WJ, Johnson WR, Duncan GA, Huckins G, Heed WB. 1999. Ecological
genetics of cactophilic Drosophila. In: Robichaux R, editor. Ecology of
Sonoran Desert plants and plant communities. Tucson: University of
Arizona Press. p. 164–214.
Etges WJ, Klassen CS. 1989. Influences of atmospheric ethanol on adult
Drosophila mojavensis: altered metabolic rates and increases in fitness among
populations. Physiol Zool. 62:170–193.
Etges WJ, Over KF, de Oliveira CC, Ritchie MG. 2006. Inheritance of courtship song variation among geographically isolated populations of Drosophila
mojavensis. Anim Behav. 71:1205–1214.
Coyne JA, Orr HA. 2004. Speciation. Sunderland (MA): Sinauer.
Feder JL, Egan SP, Nosil P. 2012. The genomics of speciation-with-geneflow. Trends Genet. 28:342–350.
Dambroski HR, Linn C Jr, Berlocher SH, Forbes AA, Roelofs W, Feder JL.
2005. The genetic basis for fruit odor discrimination in Rhagoletis flies and its
significance for sympatric host shifts. Evolution. 59:1953–1964.
Feder JL, Nosil P. 2009. Chromosomal inversions and species differences:
when are genes affecting adaptive divergence and reproductive isolation
expected to reside within inversions? Evolution. 63:3061–3075.
Dice LR, Blossom PM. 1937. Studies of mammalian ecology in sourthwestern North America with special attention to the colors of desert mammals.
Carnegie Inst Wash Publ. 485:1–129.
Feder JL, Xie X, Rull J, Velez S, Forbes A, Leung B, Dambroski H, Filchak
KE, Aluja M. 2005. Mayr, Dobzhansky, and Bush and the complexities
of sympatric speciation in Rhagoletis. Proc Natl Acad Sci USA. 102(Suppl
1):6573–6580.
Doi M, Matsuda M, Tomaru M, Matsubayashi H, Oguma Y. 2001. A locus
for female discrimination behavior causing sexual isolation in Drosophila.
Proc Natl Acad Sci USA. 98:6714–6719.
Fellows DP, Heed WB. 1972. Factors affecting host plant selection in desertadapted cactiphilic Drosophila. Ecology. 53:850–858.
767
Journal of Heredity
Fogleman JC, Danielson PB, MacIntyre RJ. 1998. The molecular basis of
adaptation in Drosophila: the role of cytochrome P450s. Evol Biol. 30:15–77.
genetics and biology of Drosophila, Vol. 3e. New York: Academic Press. p.
311–345.
Fogleman JC, Foster JLM. 1989. Microbial colonization of injured cactus
tissue Stenocereus gummosus and its relationship to the ecology of cactophilic
Drosophila mojavensis. App Environ Microbiol. 55:100–105.
Howard RW, Blomquist GJ. 1982. Chemical ecology and biochemistry of
insect hydrocarbons. Ann. Rev. Entomol. 27:149–172.
Ford EB. 1975. Ecological genetics. New York: Wiley and Sons.
Howard RW, Blomquist GJ. 2005. Ecological, behavioral, and biochemical
aspects of insect hydrocarbons. Annu Rev Entomol. 50:371–393.
Frentiu FD, Adamski M, McGraw EA, Blows MW, Chenoweth SF. 2009. An
expressed sequence tag (EST) library for Drosophila serrata, a model system
for sexual selection and climatic adaptation studies. BMC Genomics. 10:40.
Huang da W, Sherman BT, Lempicki RA. 2009. Systematic and integrative
analysis of large gene lists using DAVID bioinformatics resources. Nat
Protoc. 4:44–57.
Funk DJ, Filchak KE, Feder JL. 2002. Herbivorous insects: model systems
for the comparative study of speciation ecology. Genetica. 116:251–267.
Jennings JH, Etges WJ. 2010. Species hybrids in the laboratory but not in
nature: a reanalysis of premating isolation between Drosophila arizonae and
D. mojavensis. Evolution. 64:587–598.
Galindo K, Smith DP. 2001. A large family of divergent Drosophila odorantbinding proteins expressed in gustatory and olfactory sensilla. Genetics.
159:1059–1072.
Galvan A, Ioannidis JP, Dragani TA. 2010. Beyond genome-wide association studies: genetic heterogeneity and individual predisposition to cancer.
Trends Genet. 26:132–141.
Gastil RG, Phillips RP, Allison EC. 1975. Reconnaissance geology of the
state of Baja California. Geol Soc Amer Mem. 140:1–170.
Gerhardt HC, Huber F. 2002 Acoustic communication in insects and anurans. Chicago: Univ. Chicago Press.
Gibson G. 2008. The environmental contribution to gene expression profiles. Nat Rev Genet. 9:575–581.
Gleason JM, Jallon JM, Rouault JD, Ritchie MG. 2005. Quantitative trait
loci for cuticular hydrocarbons associated with sexual isolation between
Drosophila simulans and D. sechellia. Genetics. 171:1789–1798.
Gleason JM, James RA, Wicker-Thomas C, Ritchie MG. 2009. Identification
of quantitative trait loci function through analysis of multiple cuticular
hydrocarbons differing between Drosophila simulans and Drosophila sechellia
females. Heredity (Edinb). 103:416–424.
Gleason JM, Nuzhdin SV, Ritchie MG. 2002. Quantitative trait loci affecting
a courtship signal in Drosophila melanogaster. Heredity (Edinb). 89:1–6.
Gompert Z, Comeault AA, Farkas TE, Feder JL, Parchman TL, Buerkle CA,
Nosil P. 2014. Experimental evidence for ecological selection on genome
variation in the wild. Ecol Lett. 17:369–379.
Kang EY, Han B, Furlotte N, Joo JW, Shih D, Davis RC, Lusis AJ, Eskin E.
2014. Meta-analysis identifies gene-by-environment interactions as demonstrated in a study of 4,965 mice. PLoS Genet. 10:e1004022.
Kaufman D. 1974. Adaptive coloration in Peromyscus polionotus: experimental
selection by owls. J Mammal. 55:271–283.
Knowles LL, Markow TA. 2001. Sexually antagonistic coevolution of a
postmating-prezygotic reproductive character in desert Drosophila. Proc Natl
Acad Sci USA. 98:8692–8696.
Leips J, Mackay TF. 2000. Quantitative trait loci for life span in Drosophila melanogaster: interactions with genetic background and larval density. Genetics.
155:1773–1788.
Lewontin RC. 1991. Biology as ideology: The doctrine of DNA. New York:
HarperCollins.
Lewontin RC. 2000. The triple helix: gene, organism, and environment.
Cambridge: Harvard University Press.
Linnen CR, Poh YP, Peterson BK, Barrett RD, Larson JG, Jensen JD,
Hoekstra HE. 2013. Adaptive evolution of multiple traits through multiple
mutations at a single gene. Science. 339:1312–1316.
Machado CA, Matzkin LM, Reed LK, Markow TA. 2007. Multilocus nuclear
sequences reveal intra- and interspecific relationships among chromosomally polymorphic species of cactophilic Drosophila. Mol Ecol. 16:3009–3024.
Mackay TFC. 2014. Epistasis and quantitative traits: using model organisms
to study gene-gene interactions. Nat Rev Genet. 15:22–33.
Greenfield MD. 2002. Signalers and receivers: mechanisms and evolution of
arthropod communication. New York: Oxford Univ. Press.
Mackay TFC, Anholt RRH. 2007. Ain’t misbehavin’? Genotype-environment
interactions and the genetics of behavior. Trends in Genet. 23:311–314.
Greenspan RJ, Ferveur JF. 2000. Courtship in Drosophila. Annu Rev Genet.
34:205–232.
Majerus MEN. 1998. Melanism: evolution in action. Oxford: Oxford
University Press.
Grishkevich V, Yanai I. 2013. The genomic determinants of genotype ×
environment interactions in gene expression. Trends Genet. 29:479–487.
Mallet J, Dasmahapatra KK. 2012. Hybrid zones and the speciation continuum in Heliconius butterflies. Mol Ecol. 21:5643–5645.
Gurganus MC, Fry JD, Nuzhdin SV, Pasyukova EG, Lyman RF, Mackay TF.
1998. Genotype-environment interaction at quantitative trait loci affecting
sensory bristle number in Drosophila melanogaster. Genetics. 149:1883–1898.
Malone JH, Oliver B. 2011. Microarrays, deep sequencing and the true measure of the transcriptome. BMC Biol. 9:34.
Havens JA, Orzack SH, Etges WJ. 2011. Mate choice opportunity leads
to shorter offspring development time in a desert insect. J Evol Biol.
24:1317–1324.
Hawthorne DJ, Via S. 2001. Genetic linkage of ecological specialization and
reproductive isolation in pea aphids. Nature. 412:904–907.
Heed WB. 1978. Ecology and genetics of Sonoran Desert Drosophila. In:
Brussard PF, editor. Ecological genetics: the interface. New York: SpringerVerlag. p. 109–126.
Heed WB. 1981. Central and marginal populations revisited. Dros. Inform.
Serv. 56:60–61.
Heed WB. 1982. The origin of Drosophila in the Sonoran Desert. In: Barker
JSF, Starmer WT, editors. Ecological genetics and evolution: the CactusYeast-Drosophila Model System. Sydney: Academic Press. p. 65–80.
Heed WB, Mangan RL. 1986. Community ecology of the Sonoran Desert
Drosophila. In: Ashburner M, Carson HL, Thompson JN, editors. The
768
Markow TA. 1991. Sexual isolation among populations of Drosophila mojavensis. Evolution. 45:1525–1529.
Markow TA, Fogleman JC, Heed WB. 1983. Reproductive isolation in
Sonoran Desert Drosophila. Evolution. 37:649–652.
Markow TA, Toolson EC. 1990. Temperature effects on epicuticular hydrocarbons and sexual isolation in Drosophila mojavensis. In: Barker JSF, Starmer
WT, MacIntyre RJ, editors. Ecological and evolutionary genetics of Drosophila:
monographs in evolutionary biology. New York: Plenum. p. 315–331.
Massie KR, Markow TA. 2005. Sympatry, allopatry and sexual isolation
between Drosophila mojavensis and D. arizonae. Hereditas. 142:51–55.
Matzkin LM, Eanes WF. 2003. Sequence variation of alcohol dehydrogenase
(Adh) paralogs in cactophilic Drosophila. Genetics. 163:181–194.
Mettler LE. 1957. Studies on experimental populations of Drosophila arizonensis and Drosophila mojavensis. Univ Tex Publ 5721:157–181.
Mettler LE, Nagle JJ. 1966. Corroboratory evidence for the concept of the
sympatric origin of isolating mechanisms. DIS. 41:76.
Etges • Genetics, Environments, and Speciation
Michaelson JJ, Loguercio S, Beyer A. 2009. Detection and interpretation of
expression quantitative trait loci (eQTL). Methods. 48:265–276.
SAS-Institute. 2004. SAS/STAT 9.1.2. Cary (NC): SAS Institute, Inc.
Mills LE, Batterham P, Alegre J, Starmer WT, Sullivan. DT. 1986. Molecular
genetics characterization of a locus that contains duplicate Adh genes in
Drosophila mojavensis and related species. Genetics. 112:295–310.
Schwarz D, Robertson HM, Feder JL, Varala K, Hudson ME, Ragland
GJ, Hahn DA, Berlocher SH. 2009. Sympatric ecological speciation meets
pyrosequencing: sampling the transcriptome of the apple maggot Rhagoletis
pomonella. BMC Genomics. 10:633.
Nachman MW, Hoekstra HE, D’Agostino SL. 2003. The genetic basis of
adaptive melanism in pocket mice. Proc Natl Acad Sci USA. 100:5268–5273.
Seehausen O, vanAlphen JJM, Witte F. 1997. Cichlid fish diversity threatened
by eutrophication that curbs sexual selection. Science. 277:1808–1811.
Naisbit RE, Jiggins CD, Mallet J. 2003. Mimicry: developmental genes that
contribute to speciation. Evol Dev. 5:269–280.
Shaw KL, Mullen SP. 2011. Genes versus phenotypes in the study of speciation. Genetica. 139:649–661.
Nosil P. 2012. Ecological speciation. Oxford: Oxford University Press.
Shaw KL, Parsons YM. 2002. Divergence of mate recognition behavior and
its consequences for genetic architectures of speciation. Am Nat. 159(Suppl
3):S61–S75.
Nosil P, Parchman TL, Feder JL, Gompert Z. 2012. Do highly divergent
loci reside in genomic regions affecting reproductive isolation? A test using
next-generation sequence data in Timema stick insects. BMC Evol Biol.
12:164.
Nosil P, Vines TH, Funk DJ. 2005. Perspective: reproductive isolation caused
by natural selection against immigrants from divergent habitats. Evolution.
59:705–719.
Nuzhdin SV, Friesen ML, McIntyre LM. 2012. Genotype-phenotype mapping in a post-GWAS world. Trends Genet. 28:421–426.
Oliveira DC, Almeida FC, O’Grady PM, Armella MA, DeSalle R, Etges
WJ. 2012. Monophyly, divergence times, and evolution of host plant use
inferred from a revised phylogeny of the Drosophila repleta species group. Mol
Phylogenet Evol. 64:533–544.
Orr HA, Masly JP, Presgraves DC. 2004. Speciation genes. Curr Opin Genet
Dev. 14:675–679.
Oshlack A, Robinson MD, Young MD. 2010. From RNA-seq reads to differential expression results. Genome Biol. 11:220.
Pavey SA, Bernatchez L, Aubin-Horth N, Landry CR. 2012. What is needed
for next-generation ecological and evolutionary genomics? Trends Ecol
Evol. 27:673–678.
Peixoto AA, Hall JC. 1998. Analysis of temperature-sensitive mutants
reveals new genes involved in the courtship song of Drosophila. Genetics.
148:827–838.
Pritchard VL, Dimond L, Harrison JS, S Velázquez CC, Zieba JT, Burton RS,
Edmands S. 2011. Interpopulation hybridization results in widespread viability selection across the genome in Tigriopus californicus. BMC Genet. 12:54.
Quackenbush J. 2001. Computational analysis of microarray data. Nat Rev
Genet. 2:418–427.
Rajpurohit S, Oliveira CC, Etges WJ, Gibbs AG. 2013. Functional genomic
and phenotypic responses to desiccation in natural populations of a desert
drosophilid. Mol Ecol. 22:2698–2715.
Reed LK, Nyboer M, Markow TA. 2007. Evolutionary relationships of
Drosophila mojavensis geographic host races and their sister species Drosophila
arizonae. Mol Ecol. 16:1007–1022.
Rodríguez RL, Sullivan LM, Snyder RL, Cocroft RB. 2008. Host shifts and
the beginning of signal divergence. Evolution. 62:12–20.
Rosenblum EB, Hoekstra HE, Nachman MW. 2004. Adaptive reptile color
variation and the evolution of the Mc1r gene. Evolution. 58:1794–1808.
Ross CL, Markow TA. 2006. Microsatellite variation among diverging populations of Drosophila mojavensis. J Evol Biol. 19:1691–1700.
Singer MC, McBride CS. 2010. Multitrait, host-associated divergence among
sets of butterfly populations: implications for reproductive isolation and
ecological speciation. Evolution. 64:921–933.
Smith G, Fang Y, Liu X, Kenny J, Cossins AR, de Oliveira CC, Etges WJ,
Ritchie MG. 2013. Transcriptome-wide expression variation associated with
environmental plasticity and mating success in cactophilic Drosophila mojavensis. Evolution. 67:1950–1963.
Smith G, Lohse K, Etges WJ, Ritchie MG. 2012. Model-based comparisons
of phylogeographic scenarios resolve the intraspecific divergence of cactophilic Drosophila mojavensis. Mol Ecol. 21:3293–3307.
Starmer WT, Heed WB, Miranda M, Miller MW, Phaff HJ. 1976. The ecology of yeast flora associated with cactiphilic Drosophila and their host plants
in the Sonoran Desert. Microb Ecol. 3:11–30.
Starmer WT, Heed WB, Rockwood-Sluss ES. 1977. Extension of longevity
in Drosophila mojavensis by environmental ethanol: differences between subraces. Proc Nat Acad Sci USA. 74:387–391.
Stranger BE, Stahl EA, Raj T. 2011. Progress and promise of genomewide association studies for human complex trait genetics. Genetics.
187:367–383.
Taylor EB, Boughman JW, Groenenboom M, Sniatynski M, Schluter D, Gow
JL. 2006. Speciation in reverse: morphological and genetic evidence of the
collapse of a three-spined stickleback (Gasterosteus aculeatus) species pair. Mol
Ecol. 15:343–355.
Ting CT, Tsaur SC, Sun S, Browne WE, Chen YC, Patel NH, Wu CI. 2004.
Gene duplication and speciation in Drosophila: evidence from the Odysseus
locus. Proc Natl Acad Sci USA. 101:12232–12235.
Toolson EC. 1982. Effects of rearing temperature on cuticle permeability
and epicuticular lipid composition in Drosophila pseudoobscura. J Exp Zool.
222:249–253.
Turelli M, Lipkowitz JR, Brandvain Y. 2014. On the Coyne and Orr-igin
of species: effects of intrinsic postzygotic isolation, ecological differentiation, x chromosome size, and sympatry on Drosophila speciation. Evolution.
68:1176–1187.
Turner JRG. 1977. Butterfly mimicry: the genetical evolution of an adaptation. Evol Biol. 10:163–206.
Tweedie S, Ashburner M, Falls K, Leyland P, McQuilton P, Marygold
S, Millburn G, Osumi-Sutherland D, Schroeder A, Seal R, et al.; FlyBase
Consortium. 2009. FlyBase: enhancing Drosophila Gene Ontology annotations. Nucleic Acids Res. 37:D555–D559.
Ruiz A, Heed WB, Wasserman M. 1990. Evolution of the mojavensis cluster of
cactophilic Drosophila with descriptions of two new species. J Hered. 81:30–42.
Van Devender TR. 1990. Late quarternary vegetation and climate of
the Sonoran Desert, United States and Mexico. In: Betancourt JL, Van
Devender TR, Martin PS, editors. Packrat middens: the last 40,000 years of
biotic change. Tucson: Univ. Arizona Press. p. 134–165.
Ryner LC, Goodwin SF, Castrillon DH, Anand A, Villella A, Baker
BS, Hall JC, Taylor BJ, Wasserman SA. 1996. Control of male sexual
behavior and sexual orientation in Drosophila by the fruitless gene. Cell.
87:1079–1089.
Wasserman M. 1962. Cytological studies of the repleta group of the genus
Drosophila. V. The mulleri subgroup. Univ Texas Publ. 6205:85–117.
Ruiz A, Heed WB. 1988. Host-plant specificity in the cactophilic Drosophila
mulleri species complex. J Anim Ecol. 57:237–249.
Via S. 2009. Natural selection in action during speciation. Proc Natl Acad Sci
USA. 106(Suppl 1):9939–9946.
769
Journal of Heredity
Wasserman M. 1992. Cytological evolution of the Drosophila repleta species
group. In: Krimbas CB, Powell JR, editors. Drosophila inversion polymorphism. Boca Raton: CRC Press. p. 455–552.
Yukilevich R. 2014. The rate test of speciation: estimating the likelihood
of non-allopatric speciation from reproductive isolation rates in Drosophila.
Evolution. 68:1150–1162.
Wasserman M, Koepfer HR. 1977. Character displacement for sexual isolation between Drosophila mojavensis and Drosophila arizonensis. Evolution.
31:812–823.
Zouros E, d’Entremont CJ. 1974. Sexual isolation among populations of
Drosophila mojavensis race B. Dros Inf Serv. 51:112.
Zouros E, d’Entremont CJ. 1980. Sexual isolation among populations of
Drosophila mojavensis: response to pressure from a related species. Evolution.
34:421–430.
Williams GC. 1966. Adaptation and natural selection: a critique of some
current evolutionary thought. Princeton (NJ): Princeton University Press.
Wray GA. 2013. Genomics and the evolution of phenotypic traits. Ann Rev
Ecol Evol Syst. 44:51–72.
Young DJ, Vacek DC, Heed WB. 1981. The facultatively anaerobic bacteria
as a source of alcohols in three breeding substrates of cactophilic Drosophila.
Dros Infor Serv. 56:165–166.
770
Received January 16, 2014; First decision March 5, 2014;
Accepted May 31, 2014
Corresponding Editor: Kerry Shaw