Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Evolutionary mechanisms responsible for genetic and phenotypic variation
(USC Thesis Other)
Evolutionary mechanisms responsible for genetic and phenotypic variation
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
Evolutionary mechanisms responsible for
genetic and phenotypic variation
by
Wendy T Vu
________________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Molecular Biology)
August 2015
Copyright 2015 Wendy T Vu
ii
Acknowledgements
During my graduate program, I was fortunate enough to have the support and guidance
from team of talented and inspirational scientist whom have shaped me into the scientist I
am today. My scientific career began when I joined Sergey Nuzhdin’s lab as an
undergraduate researcher at the University of California, Davis. Sergey has played a pivotal
role in my growth as a scientist. He introduced me to a wide range of scientific topics and
provided me with the freedom to explore my scientific passions. I have had the opportunity
to travel the world and collaborate with scientist from Tunisia, Portugal, India and Russia.
And along the way, l learnt invaluable lessons about what it takes to be a researcher and the
importance of scientific collaboration. For these reasons, I am thankful to have had Sergey
as my graduate advisor. I would like to thank Maren Friesen for taking the time to mentor
me during my graduate career. Through her, I have learnt what it takes to be a leader and
manage a team of scientist. I would also like to thank Peter Chang, who is both a mentor
and a friend, for introducing me to the world of computer science and bioinformatics. He
has played a key role in shaping me into a more dynamic and resourceful scientist. I would
like to thank Hamdi Kitapci for always taking the time to help me with technical computer
science problems; it has been a joy collaborating with you. And last but not least, I would
like to extend a thank you to Brad Foley. He is both my mentor and long time friend. He
played a pivotal role in guiding me through the struggles of scientific growth and has always
been a major source of inspiration.
Last but not least, I would like to thank my family and friends. They have played a critical
role in shaping me into the person I am today. I would like to thank my mom, Diana Vu, for
raising me as a single parent and showing me what it takes to be a strong and independent
woman. I want to also thank my best friends, Aysen Erdem and Julia Babamuratova, for
their support during my graduate career. Lastly, I want to thank my partner, Felix Fricke, for
providing a lifetime of laughter and inspiration. Through you, I can see my true self.
iii
Table of Contents
Acknowledgements ............................................................................................................................... ii
List of Tables ........................................................................................................................................ vi
List of Figures ...................................................................................................................................... vii
Preface .................................................................................................................................................. viii
Abstract .................................................................................................................................................. ix
Chapter 1: Introduction ..................................................................................................................... 1
1.1 Natural selection and genetic variation ................................................................................................ 1
1.2 Transposable elements: a source of genetic and phenotypic variation ........................................... 2
1.2.1 Mechanisms responsible for TE activity and copy number control ....................................... 5
1.3 Maintenance of genetic and phenotypic variation ............................................................................. 7
1.3.1 Evolutionary consequences of self-fertilization .......................................................................... 8
1.3.2 Genotype-environment interaction ............................................................................................. 10
1.4 References .............................................................................................................................................. 13
Chapter 2: Genetic variation of transposable element host control mechanism ....................22
2.1 Abstract .................................................................................................................................................. 22
2.2 Introduction ........................................................................................................................................... 23
2.3 Materials and Methods ......................................................................................................................... 26
2.3.1 Wild-type Drosophila melanogaster lines and copia transpositions ............................................... 26
2.3.2 Amplification of copia plasmid from parental lines Ore and 2b .............................................. 27
2.3.3 Electron microscopy images of VLPs in testes ......................................................................... 29
2.3.4 Copia double LTR plasmid sequencing ....................................................................................... 30
2.3.5 Semi-quantitative analysis of concentration of copia plasmid in each RIL ............................ 31
2.3.6 QTL mapping for copia plasmid concentration ......................................................................... 32
2.3.7 Deficiency mapping of copia plasmid concentration ................................................................. 33
2.4 Results ..................................................................................................................................................... 36
2.4.1 Copia plasmid in parental stocks .................................................................................................. 36
2.4.2 Electron microscopy images of VLPs in testes ......................................................................... 37
2.4.3 Copia plasmid in RILs .................................................................................................................... 38
2.4.4 QTL mapping of copia plasmid concentration ........................................................................... 41
2.4.5 Deficiency mapping of copia plasmid concentration ................................................................. 42
2.5 Discussion .............................................................................................................................................. 46
2.6 References .............................................................................................................................................. 50
iv
Chapter 3: Maintenance of genetic variation and microgeographical adaptation ...................55
3.1 Abstract .................................................................................................................................................. 55
3.2 Introduction ........................................................................................................................................... 56
3.3 Results ..................................................................................................................................................... 61
3.3.1 Absence of genotype-environment interactions: flowering time is fixed and performance
traits are plastic ......................................................................................................................................... 61
3.3.2 Topographical variation influences soil salinity variability ...................................................... 63
3.3.3 Flowering time varies along the soil salinity gradient in the field ........................................... 66
3.3.4 Genome sequencing and population structure analysis ........................................................... 68
3.3.5 Selection along microenvironmental gradients .......................................................................... 69
3.4 Discussion .............................................................................................................................................. 73
3.5 Materials and Methods ......................................................................................................................... 78
3.5.1 Field collection and greenhouse experiment ............................................................................. 78
3.5.2 Genome sequencing and library construction ........................................................................... 78
3.5.3 Population structure analysis ........................................................................................................ 79
3.5.4 Statistical analysis on plant traits .................................................................................................. 80
3.5.5 Fst estimates of 10kb windows .................................................................................................... 80
3.5.6 Multiple regression on allele frequency ...................................................................................... 80
3.6 References .............................................................................................................................................. 81
Chapter 4: Genetic variation in transgenerational plasticity of the seed transcriptome .........86
4.1 Abstract .................................................................................................................................................. 86
4.2 Introduction ........................................................................................................................................... 88
4.3 Materials and Methods ......................................................................................................................... 93
4.3.1 Parental and offspring environment ........................................................................................... 93
4.3.2 Transcriptome library construction ............................................................................................. 94
4.3.3 Mapping and normalization of sequencing reads ..................................................................... 95
4.3.4 Offspring phenotype data analysis .............................................................................................. 95
4.3.5 Seed transcriptome data analysis ................................................................................................. 96
4.3.6 Network and functional analysis .................................................................................................. 96
4.4 Results ..................................................................................................................................................... 97
4.4.1 Sequencing stored seed transcriptome ........................................................................................ 98
4.4.2 Stored seed transcripts are annotated to be involved in germination and dormancy
processes ................................................................................................................................................. 100
4.4.3 Genotype-dependent parental environmental effects on stored seed transcripts ............. 100
4.4.4 Genotype-dependent transgenerational plasticity on germination behavior ...................... 101
4.4.5 Genotype-dependent parental environmental effects on seed size and the absence of seed
size effects on germination behavior .................................................................................................. 105
4.4.6 Salt responsive mature seed transcripts are involved in dormancy and ABA-related
processes ................................................................................................................................................. 106
4.5 Discussion ............................................................................................................................................ 111
v
4.6 References ............................................................................................................................................ 120
Appendix A ...................................................................................................................................... 128
List of Medicago truncatula and Arabidopsis orthologs involved in TN1.13 and TN1.15 transcriptional
salt response.
Appendix B ...................................................................................................................................... 134
The ecological genomic basis of salinity adaptation in Tunisian Medicago truncatula
Appendix C ...................................................................................................................................... 135
Salinity Adaptation and the contribution of parental effects in Medicgao truncatula
vi
List of Tables
2.1 Deficiency effects of copia plasmid concentrations ..................................................................29
2.2 Copia plasmid LTR junction Primers and Probes ....................................................................31
3.1 Analysis of variance on vegetative and reproductive traits ....................................................62
3.2 Pearson correlation coefficients (r) for early and vegetative growth traits with respect to
seed mass and flowering time ............................................................................................................63
3.3 Annotations of genes associated with significant SNPs that correlate with elevation in the
field along with corresponding SNP effects ...................................................................................71
4.1 Complete list of significant gene ontology terms for total genes expressed in the seed
transcriptome enriched in biological pathways ...............................................................................99
4.2 ANOVA table of P-values for offspring traits ...................................................................... 103
4.3 ANOVA table of F-values for offspring germination timing ............................................. 103
4.4 Genes associated with ABA up-regulation within the dormancy coexpression sub-
network .............................................................................................................................................. 111
4.5 Pearson correlation coefficients for seed size in relation to leaf size and leaf counts for all
four genotypes. ................................................................................................................................. 113
vii
List of Figures
2.1 Copia transposition process .........................................................................................................26
2.2 Copia plasmids in parental lines ...................................................................................................37
2.3 EM micrographs of testes-cross section parental lines ...........................................................38
2.4 Distribution of copia plasmid concentrations among parental and recombinant inbred
lines .......................................................................................................................................................39
2.5 Correlations of plasmid concentration with copy number and transcript level of copia
element ..................................................................................................................................................41
2.6 Location of QTL for copia plasmid concentration with respect to transcript level and
transposition rate .................................................................................................................................42
2.7 Chromosome position of deficiency ..........................................................................................43
3.1 Plot of plant positions sampled along the latitudinal and longitudinal axis .........................64
3.2 Distribution of pairwise spatial distances of sampled ecotypes in meters (m) ....................65
3.3 3D spatial distribution of accessions collected from the field and corresponding salinity
levels ......................................................................................................................................................66
3.4 Correlation between flowering time and soil salinity ..............................................................67
3.5 Correlation between flowering time and elevation ..................................................................68
3.6 STRUCTURE plot comparing the Soliman population with four previously characterized
Tunisian populations ..........................................................................................................................69
3.7 Manhattan plots of the SNPs that are significantly associated with latitude, longitude,
elevation and soil conductivity in the field ......................................................................................70
3.8 Distribution of average F
st
values ...............................................................................................73
4.1 Norms of reaction plots for germination timing in response to parental and offspring
environment ...................................................................................................................................... 104
4.2 Seed weight comparison of genotypes developing in 0mM and 100 mM parental NaCl
conditions .......................................................................................................................................... 106
4.3 SeedNet coexpression networks topology ............................................................................. 108
4.4 Discrete tightly clustered modules representing putative biological pathways ................ 110
viii
Preface
This dissertation consists of a section describing research in Drosophila melanogaster and two
sections outlining work done in Medicago truncatula. The experiments and analysis performed
here at USC has resulted in several published and prepared manuscripts as listed below:
Vu WT, and Nuzhdin SV. "Genetic variation of copia suppression in Drosophila
melanogaster." Heredity 106.2 (2010): 207-217.
Vu WT, Chang PL, Moriuchi KS, Friesen ML. “Genetic variation in transgenerational
plasticity of the seed transcriptome and offspring germination response to salinity stress in
Medicago truncatula.” BMC Evolutionary Biology (2015), accepted for publication, in press.
Vu WT, Chang PL, von Wettberg EJB, Friesen ML, Kitapci TH, Nuzhdin SV. “Patterns of
microgeographical adaptation to a patchy saline environment in Medicago truncatula.” Plos
Biology, in prep.
Friesen ML, von Wettberg EJB, Badri M, Moriuchi KS, Barhoumi F, Chang PL, Cuellar-
Ortiz S, Cordeiro MA, Vu WT, Arraouadi S, Djebali N, Zribi K, Badri Y, Porter SS, Aouani
ME, Cook DR, Strauss SY, Nuzhdin SV. “The ecological basis of salinity adaptation in
Tunisian Medicago truncatula.” BMC Genomics (2015), accepted for publication, in press.
Moriuchi KS,
Friesen ML, Cordeiro MA, Badri M, Vu WT, Main BJ, Elarbi Aouani M,
Nuzhdin SV, Strauss SS, and von Wettberg EJB. “Salinity adaptation and the contribution of
parental effects in Medicago truncatula.” American Journal of Botany (2015), in review.
Wendy Vu
Los Angeles, California, USA
August 2015
ix
Abstract
Genetic variation is essential for natural selection to operate and facilitate evolutionary
processes. Identifying the mechanisms that influence the degree of genetic variation is
important for our understanding of population differences and species diversity.
Mechanisms that produce genetic variation are those that generate mutations in nucleotide
sequences; such mechanisms include replication error, DNA damage and insertions and
deletions caused by transposable element transpositions. A portion of my dissertation work
will focus on transposable elements, which are mobile genetic elements that replicate by
inserting themselves into the host genome. Like mutations caused by replication error or
DNA damage, these insertions are often deleterious; however, in some cases can lead to new
genetic variants contributing to phenotypic variation. Although the strength and
effectiveness of natural selection is proportional to the amount of genetic variation, high
rates of mutation is detrimental and organisms have evolved mechanisms to prevent
mutations from occurring. Chapter 2 examines the coevolution of transposable
elements (TE) and their host to identify mechanisms that prevent transpositions and
control TE copy number. This study focuses on genetic variation and the population
dynamics of TE host silencing pathways, specifically the copia long-terminal-repeat
retrotransposon in Drosophila melanogaster.
Reduced genetic variation is an inevitable consequence of natural selection, but studies in
both morphological and physiological traits have detected considerable amounts of genetic
and phenotypic variation within and among populations and species. This indicates that
there are evolutionary mechanisms that work to maintain genetic and phenotypic variation.
x
Differential selection due to spatial and temporal environmental variation can lead to the
maintenance of genetic and phenotypic variation. Genetic differentiation is often seen in
populations isolated by large-scale geographical distances; however, population
differentiation can also occur within smaller geographical distances, where
microenvironmental variability within a single habitat can facilitate differential selection.
Chapter 3 focuses on the maintenance of genetic and phenotypic variation in a
natural plant population adapted to a heterogeneous environment. This study
examines local adaptation of a single self-fertilizing population of an emerging model
legume, Medicago truncatula, and identifies patterns of microgeographcial adaptation to a saline
environment exhibiting high variability in soil salinity levels.
Adaptation to spatial and temporal environmental variability has been hypothesized to favor
the evolution of phenotypic plasticity. Phenotypic plasticity is defined as a single genotype
expressing alternative phenotypes in response to different environments or environmental
cues. This type of phenotypic flexibility allows organisms to adapt to climate change and
inhabit a broader range of environments. Phenotypic plasticity is especially important for
sessile organisms that have little choice in their growth conditions. Plants, for instance, rely
on environmental cues to make developmental transitions as well as adaptively adjust
physiological and developmental traits in response to abiotic and biotic stress. In order for
plasticity to evolve, there must be genotypic differences in phenotypic plasticity—this kind
of genetic variation is known as genotype-environment interactions. In addition to the
immediate environment, the parental environment also plays a pivotal role in influencing
plastic plant growth and development. This form of plasticity is known as transgenerational
plasticity, where cues from the parental environment are transmitted to influence offspring
xi
phenotype. Transgenerational plasticity contributes to phenotypic variation and can
influence the strength of selection and rate of adaptation. In order for transgenerational
plasticity to itself evolve, it must be heritable and exhibit variation among genotypes.
Chapter 4 test for genetic variation in transgenerational plasticity of natural Medicago
truncatula genotypes and identifies novel molecular mechanisms that are influenced
by the parental environment to facilitate changes in offspring development and
response. This study examines parental exposure to salinity stress and test whether the
parental environment influences offspring germination behavior and the expression of
stored seed transcripts (necessary for seed germination) to infer biological pathways
mediating transgenerational plastic offspring germination response.
1
Chapter 1
Introduction
1.1 Natural selection and genetic variation
Natural selection is a non-random process that is key to facilitating evolutionary change. It is
defined as a gradual process, where heritable biological traits become more or less common
in a population and this is based on the organism’s reproductive success in a particular
environment. Genetic variation is what fuels natural selection and has been a subject of great
interest in both theoretical (Gillespie, 1984; Lande; Charlesworth & Charlesworth; Hartl &
Clark, 1975; Charlesworth & Hughes, 2000; Nei, 1987) and empirical (Hamrick et al., 1979;
Rendel, 1943; Lack, 1966; Allard & Jain, 1962; Radwan, 2008; Byers, 2005) studies in
evolutionary biology. Genetic variability is a direct consequence of the accumulation of non-
lethal mutations and these mutations arise randomly from a number of different processes
that include replication error, DNA damage, insertions and deletions caused by either
transposable element transpositions or replication slippage (Bertram, 2000; Aminetzach et al.
2005; Burrus & Walder, 2004). In general, mutations are classified as deleterious, beneficial
or neutral, and this is based on the influence that the mutations have on an organism’s
fitness in a given environment. Deleterious and beneficial mutations are genetic variants that
influence traits related to fitness and are, therefore, under selection. In contrast, neutral
mutations have no differential effect on fitness; therefore, exhibit little to no selective
pressure.
2
Experimental studies have demonstrated that natural selection favors intermediate optimum
phenotypes in natural populations and this kind of stabilizing selection has been
characterized in many quantitative traits (Hamrick et al., 1979; Rendel, 1943; Lack, 1966;
Allard & Jain, 1962; Radwan, 2008; Byers, 2005). Many physiological, morphological and
behavioral traits are typically under stabilizing selection (Rendel, 1943) and theoretical
models have demonstrated that quantitative traits under stabilizing selection will exhibit a
reduction of genetic variability (Fisher, 1930; Robertson, 1965; Lewontin, 1964b), and this
has been confirmed in numerous experimental studies (Waddington, 1960; Scharloo, 1964;
Gibson & Bradley, 1974). The reduction of genetic variability will slow and limit the
response of selection; therefore, the loss of genetic variation as a consequence of natural
selection must be balanced by mechanisms that either maintain or introduce genetic
diversity. There are a number of empirical studies in both morphological and physiological
traits that have detected a considerable amount of genetic and phenotypic variation within
and between populations and species (Hamrick et al. 1996; Hamrick et al., 1992; Petri et al.,
1998; Violle et al., 2012), indicating that there are evolutionary mechanisms, apart from
mutations, that work to maintain diversity. Factors that maintain genetic variability has been
one of the most debated topics in evolutionary biology (Hedrick et al., 1976; Gillespie and
Turelli, 1989; Lynch and Walsh, 1998; Barton and Keightley, 2002). My thesis identifies the
mechanisms that contribute to and maintain genetic and phenotypic variation, as well as
genetic variants that have the propensity to cause adaptive responses in both plants and
animals.
3
1.2 Transposable Elements: a source of genetic and phenotypic variation
Transposable elements (TE’s) are discrete mobile DNA segments that insert themselves into
the host genome (Kidwell and Lisch, 1997). Barbara McClintock discovered TE’s ~60 years
ago and called it “jumping genes” and studies have since found that TE’s are ubiquitous in
most living organisms and comprise a significant portion of animal and plant genomes
(McClintock, 1950). TE’s are estimated to account for up to 90% of the genome in some
plants (Finnegan, 1992; Sanmiguel and Bennetzen, 1998) and make up more than 50% of the
human genome (Feschotte and Pritham, 2007)and an estimated ~10-15% of the Drosophila
genome (Pimpinelli et al., 1995). Like spontaneous mutations, TE induced mutations are
random and primarily deleterious to their host (Pasyukova et al., 2004). Most TE insertions
are postulated to reduce fitness by causing chromosomal breakage (Brookfield, 1991),
disrupting gene expression and inducing ectopic exchange (Charlesworth and Langley, 1986;
Langley et al., 1988; Charlesworth et al., 1994). Studies analyzing frequency of insertion sites
have indirectly demonstrated that TE insertions are weakly selected against (Charlesworth
and Langley, 1989; Biémont et al., 1990; Charlesworth et al., 1994). If TE’s have such a
deleterious effect on the host genome and fitness, why does this relationship still persist?
There has been a long-standing debate on whether TE’s can introduce functional genetic
variation into the host genome that play a role in adaptive responses (McCLINTOCK, 1950;
Charlesworth et al., 1994; Kidwell and Lisch, 1997; Oliver and Greene, 2012).
Barbara McClintock first proposed the possibility that TE’s may play a role in stress
response in maize (McClintock, 1950; 1956), which sparked a series of studies testing the
significance of TE’s in adaptive stress responses (Chung et al., 2007; Darboux et al., 2007;
González et al., 2009; 2010; Schmidt et al., 2010). It has been hypothesized that transposons
4
are the driving force of genome evolution in both plants and animals (Kazazian, 2004) and
even the evolution of mating systems in plants (Gerashchenkov and Rozhnova, 2010). In
plants, TE’s have been found to be a source of rapid genome size expansion and may
contribute to genome size variation among Arabidopsis species (Hu et al., 2011). Recent
studies have demonstrated that TE’s may play a key role in both neurodevelopment and
disease in humans (Reilly et al., 2013). With the advances in genomic sequencing,
neurobiologists are beginning to understand the role of TE’s in the evolution of neuronal
function and are now considering them as a viable source of genetic variation.
Although mutation is one of the primary drivers of evolution on both the genomic and
phenotypic level, TE-induced genetic changes (rather than replication error) have been
proposed to be the primary force of gene expression (Kidwell and Lisch, 1997) and protein
evolution (Shapiro, 2010). Complex rearrangements induced by transposons near promoter
regions have been shown to produce new variants that influence tissue specificity of RNA
expression in Drosophila (Kloeckener-Gruissem and Freeling, 1995). A classic example of P-
element insertion into the exonic region of the Drosophila eye pigmentation gene resulted in
the white-eyed phenotype (Rubin and Spralding, 1982), thus potentially contributing to
phenotypic variation. Furthermore, TE induced inversions in Aratirrhinum have introduced
a series of genome rearrangements near the anthocyanine pigmentation gene, resulting in
new variations of flower color patterning in the flower tube (Lister et al., 1993). Controlled
selection experiments in Drosophila have demonstrated that TE induced genetic variation in
quantitative traits (Mackay, 1989; Mackay et al., 1992), where TE transpositions increased
quantitative variability in bristle numbers 30 times greater than what is expected from
spontaneous mutations. And numerous studies in inbred mouse strains have shown
5
transposon generated polymorphic traits, such as coat color variation (King et al., 1986;
Duhl et al., 1994; Yamada et al., 2006; Li et al., 2010). Although all these examples above are
of TE-induced genetic and phenotypic variability in lab strains, these examples are merely
anecdotal, but they do give us an idea of the potential of TE’s to introduce new genetic
variation that can contribute to phenotypic variation.
1.2.1 Mechanisms responsible for TE activity and copy number control
Because the accumulation of TE’s are associated with a decrease in fitness (Pasyukova et al.,
2004), there is a general agreement among theoretical and empirical studies that TE activity
and copy number must be controlled and these studies have focused on understanding the
dynamics of host-TE coevolution identify potential mechanisms maintaining this
relationship (Montgomery et al., 1987; Langley et al., 1988; Biémont et al., 1990;
Charlesworth et al., 1994; Nuzhdin et al., 1997; Pasyukova et al., 2004). While some TE-
induced mutations are beneficial, it is clear that their net effect on host fitness is negative,
with an estimated fitness cost in excess of 5% (Charlesworth et al., 1994). How have they
continued to parasitize the genome without self-destructing by means of killing the host?
One explanation seems to be that TE activity resulting from copy number increase is
balanced by selection against hosts with TE induced deleterious effects. Much like error
correcting mechanisms implemented to control the deleterious effect high rates of mutations
that arise from DNA replication error, molecular mechanisms have evolved to control the
rate of TE transpositions.
There are approximately 50 families of TEs that populate the genome of Drosophila
melanogaster, a model species for studying the dynamic relationship between TEs and their
6
host. Horizontal transmission is thought to be the primary mode of TE infection across
species and populations. P-elements, for instance, have recently invaded the genome of D.
melanogaster and have managed to rapidly sweep through the population within the last 50
years (Brookfield, 1986). With such sweeps, the number of elements within a host might
continue to increase and will do so at the expense of host fitness (Charlesworth and Langley,
1989; Mackay et al., 1992; Houle and Nuzhdin, 2004; Biémont and Vieira, 2006). Since TE
survival is dependent on the host survival and reproduction, a decrease in host fitness will
inevitably interfere with the chances of TEs propagating through the population. This
dynamic power play between TEs and their hosts has been shown to ignite co-evolution
between these players. Therefore a self-imposed control mechanism is likely to evolve in
TEs to prevent detrimental fitness affects to the host. Indeed, some elements such as Type I
transposons, can restrict their transposition at high copy number (Biémont et al., 1990; Lohe
et al., 1995). However, self-regulation appears to not be an evolutionarily stable strategy for
retrovirus-like elements (Brookfield, 1991). For instance, copia and Doc elements do not
show reduced transpositions when TE copy number increase (Nuzhdin et al., 1997).
Host fitness loss can be partially rescued if host genes controlling transpositions sweep
through the infected population. However, these sweeps might not result in fixation of the
host transposition suppressors but are likely to remain incomplete due to weak selection
(Nuzhdin et al., 1997), which would then result in an overall low transposition rate observed
in nature. This hypothesis stems from the fact that D. melangaster genotypes differ in which
TE families are active. One genotype, for example, showed transposition of hobo and I
elements but none for copia and 412 families (Harada et al., 1990). In another genotype,
copia, 297, and FB4 were active but I, mdg1, gypsy, hobo, 2161, 2242, and Doc elements
7
were not (Eggleston et al., 1988). In yet another genotype, out of 13 TE families, only copia,
I, roo and Doc were found to be actively transposing (Nuzhdin and Mackay, 1995). These
studies collectively suggest that there exists natural variation in TE stability and that the host
genetic background is an important factor that influences TE activity and thus copy number
maintenance in natural populations. In chapter 2 of my thesis, I explore genetic variation of host control
mechanisms that keep TE activity low in germ cells of Drosophila melanogaster and identify a candidate
mechanism responsible for controlling transposition of a retro-transposon, Copia.
1.3 Maintenance of genetic and phenotypic variation
The details of how polymorphisms are maintained in wild populations are not well
understood (Turelli and Barton, 2004; Mitchell-Olds et al., 2007). Although it has been
hypothesized that the balance between selection and mutation could potentially maintain
genetic variation, this would require very weak selection and high mutation rates with large
number of loci affecting the trait under selection; however, this is not a likely scenario given
the current estimates of mutation rates (Bulmer, 1989; Bulmer, 1991). Balancing selection is
an evolutionary process that can lead to the maintenance of genetic variation (Gillespie and
Turelli, 1989; Charlesworth, 2006; Tien et al., 2010); and with the advent of new sequencing
data, our understanding of this process on a genomic level has vastly improved. Balancing
selection is any type of selection that facilitates the maintenance of genetic variability in near
by loci linked to polymorphisms under selection (Strobeck, 1983; Hudson et al., 1987;
Nordborg, 1997). These so called “hypervariable” regions consist of neutral alleles that
contribute to standing genetic variation (Strobeck, 1983; Hudson et al., 1987; Charlesworth
et al., 1992; 1997). In addition to new mutations, standing genetic variation (neutral alleles)
allows populations to adapt to novel and changing environments, which can lead to rapid
8
evolution of adaptive phenotypes (Innan and Kim, 2004; Barrett and Schluter, 2008).
Balanced polymorphisms are typically maintained at intermediate frequencies (Charlesworth,
2006), thus contributes most to population variance of traits affecting fitness. Classic
examples of balancing selection are heterozygote advantage, frequency dependent selection,
antagonistic selection, and spatial-temporal selection of alternative alleles (Levene, 1953;
Gillespie and Turelli, 1989; Charlesworth, 2006). In chapter 3, I will examine the potential role of
balancing selection in maintaining genetic variation in a single selfing population of Medicago truncatula.
1.3.1 Evolutionary consequences of self-fertilization
In chapter 3 and chapter 4, I will examine genetic and phenotypic variation in a self-fertilizing annual
legume, Medicago trucatula. Like outcrossers, selfers produce gametes through meiosis, but
fertilization only occurs within a single hermaphroditic individual. As a result of inbreeding,
highly selfing species often exhibit lower genetic diversity (Schoen et al., 1996; Glémin and
Galtier, 2012) and are less efficient in purging deleterious mutations relative to outcrossing
species (Wright et al., 2013). Because inbreeding reduces the effective rate of recombination,
background selection becomes much stronger, and this effect is more pronounced in selfing
species (Wright et al., 2008; Glémin and Galtier, 2012). There is evidence that selfing
populations fix fewer beneficial mutations and more slightly deleterious ones (Slotte et al.,
2010). Theoretical models have demonstrated that selfing reduced effective population size
and consequently, the rate of adaptation (Pollak, 1987; Kamran-Disfani and Agrawal, 2014).
If inbreeding causes a reduction in genetic diversity and results in a fitness cost, why does
selfing still persist if it is considered an evolutionary dead end? Are there compensatory
mechanisms to offset inbreeding depression and promote the maintenance of both genetic
9
and phenotypic variation? Theoretical models suggest that increased recombination and
mutation rates may have evolved to offset the harmful effects of selfing (Holsinger, 1986;
Roze and Lenormand, 2005). Empirical studies of contrasting mating systems of primarily
selfing Arabidopsis thaliana and obligate outcrossing Arabidopsis lyrata show an increase in
recombination rates in selfing species; however, no studies to date have demonstrated high
mutation rates in self-fertilizing species. Because most mutations are deleterious and selfers
are inefficient at eliminating deleterious mutations, high mutation rates may be highly
unlikely. Interestingly, studies comparing selfing and outcrossing populations of closely
related plant species have shown that selfing populations exhibit higher phenotypic
variability than outcrossing populations (Jain and Marshall, 1967; Hillel et al., 1973; Brown
and Jain, 1979). High phenotypic variability observed in these highly selfing populations may
be a compensatory mechanism to offset the reduction of genetic variability (Bradshaw,
1965).
Environmental variability resulting in variable selective pressures can lead to the
maintenance of genetic and phenotypic variation (Via and Lande, 1985; Gillespie and Turelli,
1989; Sultan and Bazzaz, 1993). Natural selection operating in environments exhibiting
spatial and temporal heterogeneity can contribute to the maintenance of genetic variation,
where the proportion of polymorphic loci increases with the relative increase in
environmental variation (Bulmer, 1971). This can be explained by differential selection
among environmental gradients that select for alternative alleles associated with adaptive
traits, resulting in the maintenance of genetic and phenotypic variation. Although isolation
by geographical distance is generally a strong predictor of population differentiation (Wright,
1943; Loveless and Hamrick, 1984; Palumbi, 1994), there is evidence from empirical studies
10
that microgeographical variability can promote spatial-temporal selection resulting in within
population differentiation of specific loci associated with adaptation (Charbonnel and
Pemberton, 2005; Jump et al., 2006). Currently there are no recent studies to date testing
spatial-temporal selection on a whole genome scale. Chapter 3 identifies whole genome patterns of
spatial-temporal selection (a form of balancing selection) of a single selfing population of M. truncatula that is
adapted to a habitat exhibiting both spatial and temporal variability in soil salinity levels.
Theoretical and empirical studies have demonstrated that highly selfing organisms exhibit
stronger signals of balancing selection relative to outcrossing species (Nordborg, 1997;
Nordborg et al., 2005). Inbreeding causes a reduction in heterozygosity (Wright, 1921) and
reduced effective recombination rate; therefore, selfing species are inefficient at producing
new haplotype combinations (Narain, 1966). Consequently, balanced polymorphisms in
selfers are linked to larger regions of the genome that increases the propensity to maintain
more genetic variability in linked neutral sites. Linkage disequilibrium is much larger in
selfing species, spanning several thousands of nucleotides (Flint-Garcia et al., 2003;
Nordborg et al., 2005) compared to several hundreds of nucleotides in outcrossing species
(Long et al., 1998; Tenaillon et al., 2001). Because selfing species exhibit increased linkage
disequilibrium and selection on advantageous alleles will not greatly decrease genetic
variation in linked neutral sites, they are ideal organisms for studies attempting to identify
patterns of balancing selection (Nordborg, 1999).
1.3.2 Genotype-environment interaction
Selection within any population will always favor individuals that produce the most
appropriate phenotypes for a given environment. There has been a longstanding debate on
11
whether phenotypic plasticity or locally specialized or fixed phenotypes are favored in
heterogeneous environments and mix results have arisen from various studies(Levene, 1953;
Via and Lande, 1985; Gillespie and Turelli, 1989; Sultan and Spencer, 2002). Phenotypic
plasticity is defined as alternative phenotypes produced by a single genotype in response to
different environmental conditions (Bradshaw, 1965). The evolution of alternative
phenotypes is thought to be a result of localized adaptation to temporal and spatial
environmental variability thereby permitting individual genotypes to successfully grow and
reproduce in a variety of different environments. Much like the evolution of fixed
phenotypes, adaptive plasticity can only evolve in the presence of genetic differences in
phenotypic plasticity—this kind of genetic variation is known as genotype-environment
interactions (Falconer and Mackay, 1996). It has been hypothesized that genotype-
environment interaction is a major source of genetic variability in spatially and temporally
varying environments, where differential selection for alternative alleles among environments
facilitates the maintenance of genetic variation (Gillespie and Turelli, 1989). Because
plasticity contributes to phenotypic variation, it is a major factor influencing evolutionary
processes and patterns of genetic variation. (Bradshaw, 1965; Scheiner, 1998; Donohue et al.,
2001; Agrawal, 2002; Reed et al., 2010).
Transgenerational plasticity is another form of phenotypic plasticity, where the parental
environment, rather than the immediate environment, influences an individual’s
development and responses. Transgenerational plasticity also contributes to phenotypic
variation and is predicted to influence the rate of adaptation by changing the strength and
direction of responses to selection in the offspring generation (Kirkpatrick and Lande, 1989;
Hoyle and Ezard, 2012). Theoretical models have demonstrated that transgenerational
12
plasticity can be an adaptive mechanism that increases long-term fitness under
environmental heterogeneity (Kirkpatrick and Lande, 1989; Marshall, 2008; Dyer et al., 2010)
and extreme environmental shifts (Mousseau et al., 2009; Hoyle and Ezard, 2012). Parental
exposure to predation in three-spine sticklebacks and crickets have been shown to adaptively
influence offspring anti-predator behavior (Storm and Lima, 2010; Kozak and Boughman,
2012) and defense phenotypes of wild radish progeny have been correlated with parental
exposure to herbivory (Agrawal, 2002). In order for transgenerational plasticity to itself
evolve, it must be heritable and exhibit variation between genotypes. There is evidence that
transgenerational plasticity is genetically based in field studies examining the influence of the
parental environment on offspring response and performance relative to contrasting parental
environments (Lacey and Herr, 2000; Galloway and Etterson, 2007). Several studies in plants
have reported genotypic differences in adaptive transgenerational plasticity (Galloway, 2001;
Galloway and Etterson, 2007; Castro et al., 2013), with only some genotypes in a population
able to transmit adaptive environmental cues to their offspring.
Plants are an attractive model system for transgenerational plasticity because of the ease of
manipulating parental and offspring environments. Plants exhibit high levels of phenotypic
plasticity since they are sedentary organisms with little choice in their growth environment
(Bradshaw, 1965; Schlichting, 1986). Experimental manipulations showed that
transgenerational plasticity in Campanulastrum americanum is adaptive in the wild, but only
when the parental environment is a predictor of the offspring environment (Galloway and
Etterson, 2007). Because seed dispersal is often limited to the range of the parental plant's
environment (Levin and Kerster, 1974), the parental environment is likely often a good
predictor of the offspring environment, which would select for transgenerational
13
environmental cues (Mousseau et al., 2009; Plaistow and Benton, 2009; Boots and Roberts,
2012; Hoyle and Ezard, 2012; Ezard et al., 2014). Selfers are predicted to be less successful
migrants than outcrossers (Holsinger, 1986) and parent-offspring correlation is, therefore, a
necessary condition for adaptive evolution. Consistent with these expectations, study in a
Impatiens capensis demonstrated that self-fertilized offspring express higher overall fitness
when grown within close proximity to the parental plant and significantly lower fitness only
12 meters away from the parental site (Schmitt and Gamble, 1990). Chapter 4 will explore the
genotype-environment interaction in the context of transgenerational plasticity in M. truncatula, where
phenotypic plasticity is dependent on the parental environment.
1.4 References
Agrawal AA (2002). Herbivory and maternal effects: mechanisms and consequences of
transgenerational induced plant resistance. Ecology 83: 3408–3415.
Barrett R, Schluter D (2008). Adaptation from standing genetic variation. Trends in Ecology
& Evolution 23: 38–44.
Barton NH, Keightley PD (2002). Multifactorial genetic understanding quantitative genetic
variation. Nat Rev Genet 3: 11–21.
Biémont C, Vieira C (2006). Genetics: junk DNA as an evolutionary force. Nature 443: 521–
524.
Biémont C, Ronsseray S, Anxolabéhère D, Izaabel H, Gautier C (1990). Localization of P
elements, copy number regulation, and cytotype determination in Drosophila
melanogaster. Genet Res 56: 3–14.
Boots M, Roberts KE (2012). Maternal effects in disease resistance: poor maternal
environment increases offspring resistance to an insect virus. Proceedings of the Royal
Society B: Biological Sciences 279: 4009–4014.
Bradshaw AD (1965). Evolutionary significance of phenotypic plasticity in plants. Advances
in genetics.
14
Brookfield J (1986). A Model for Dna-Sequence Evolution Within Transposable Element
Families. Genetics 112: 393–407.
Brookfield J (1991). Models of Repression of Transposition in P-M Hybrid Dysgenesis by P-
Cytotype and by Zygotically Encoded Repressor Proteins. Genetics 128: 471–486.
Brown CR, Jain SK (1979). Reproductive system and pattern of genetic variation in two
Limnanthes species. Theor Appl Genet 54: 181–190.
Bulmer M (1991). The Selection-Mutation-Drift Theory of Synonymous Codon Usage.
Genetics 129: 897–907.
Bulmer MG (1971). The effect of selection on genetic variability. American Naturalist: 201–
211.
Bulmer MG (1989). Maintenance of Genetic-Variability by Mutation Selection Balance - a
Childs Guide Through the Jungle. Genome 31: 761–767.
Castro BM, Moriuchi KS, Friesen ML, Badri M, Nuzhdin SV, Strauss SY, et al. (2013).
Parental environments and interactions with conspecifics alter salinity tolerance of
offspring in the annual Medicago truncatula (P Klinkhamer, Ed.). J Ecol 101: 1281–1287.
Charbonnel N, Pemberton J (2005). A long-term genetic survey of an ungulate population
reveals balancing selection acting on MHC through spatial and temporal fluctuations in
selection. Heredity (Edinb) 95: 377–388.
Charlesworth B, Langley CH (1986). The evolution of self-regulated transposition of
transposable elements. Genetics 112: 359–383.
Charlesworth B, Langley CH (1989). The Population-Genetics of Drosophila Transposable
Elements. Annu Rev Genet 23: 251–287.
Charlesworth B, JARNE P, Assimacopoulos S (1994). The distribution of transposable
elements within and between chromosomes in a population of Drosophila
melanogaster. III. Element abundances in heterochromatin. Genet Res 64: 183–197.
Charlesworth B, Nordborg M, Charlesworth D (1997). The effects of local selection,
balanced polymorphism and background selection on equilibrium patterns of genetic
diversity in subdivided populations. Genet Res 70: 155–174.
Charlesworth D (2006). Balancing selection and its effects on sequences in nearby genome
regions. PLoS Genet 2: e64.
Charlesworth D, Morgan MT, Charlesworth B (1992). The effect of linkage and population
size on inbreeding depression due to mutational load. Genet Res 59: 49–61.
Chung H, Bogwitz MR, McCart C, Andrianopoulos A, ffrench-Constant RH, Batterham P,
et al. (2007). Cis-regulatory elements in the Accord retrotransposon result in tissue-
specific expression of the Drosophila melanogaster insecticide resistance gene Cyp6g1.
15
Genetics 175: 1071–1077.
Darboux I, Charles J-F, Pauchet Y, Warot S, Pauron D (2007). Transposon-mediated
resistance to Bacillus sphaericus in a field-evolved population of Culex pipiens (Diptera:
Culicidae). Cell Microbiol 9: 2022–2029.
Donohue K, Pyle EH, Messiqua D, Heschel MS, Schmitt J (2001). Adaptive divergence in
plasticity in natural populations of Impatiens capensis and its consequences for
performance in novel habitats. Evolution 55: 692–702.
Duhl DM, Vrieling H, Miller KA, Wolff GL, Barsh GS (1994). Neomorphic agouti
mutations in obese yellow mice. Nat Genet 8: 59–65.
Dyer AR, Brown CS, Espeland EK, McKay JK, Meimberg H, Rice KJ (2010). SYNTHESIS:
The role of adaptive trans-generational plasticity in biological invasions of plants.
Evolutionary Applications 3: 179–192.
Eggleston WB, Johnson-schlitz DM, Engels WR (1988). P-M Hybrid Dysgenesis Does Not
Mobilize Other Transposable Element Families in Drosophila-Melanogaster. Nature
331: 368–370.
Ezard THG, Prizak R, Hoyle RB (2014). The fitness costs of adaptation via phenotypic
plasticity and maternal effects (D Marshall, Ed.). Funct Ecology 28: 693–701.
Falconer DS, Mackay T (1996). I ntroduction to Q uantitative Genetics. Longman.
Feschotte C, Pritham EJ (2007). DNA transposons and the evolution of eukaryotic
genomes. Annu Rev Genet 41: 331–368.
Finnegan DJ (1992). Transposable elements. Curr Opin Genet Dev 2: 861–867.
Flint-Garcia SA, Thornsberry JM, Buckler ES (2003). Structure of linkage disequilibrium in
plants. Annu Rev Plant Biol 54: 357–374.
Galloway LF (2001). Parental environmental effects on life history in the herbaceous plant
Campanula americana. Ecology 82: 2781–2789.
Galloway LF, Etterson JR (2007). Transgenerational Plasticity Is Adaptive in the Wild.
Science 318: 1134–1136.
Gerashchenkov GA, Rozhnova NA (2010). [Mobile genetic elements in plant sex evolution].
Genetika 46: 1445–1457.
Gillespie JH, Turelli M (1989). Genotype-environment interactions and the maintenance of
polygenic variation. Genetics 121: 129–138.
Glémin S, Galtier N (2012). Genome Evolution in Outcrossing Versus Selfing Versus
Asexual Species. In: Methods in Molecular Biology, Methods in Molecular Biology.
Humana Press: Totowa, NJ Vol 855, pp 311–335.
16
González J, Karasov TL, Messer PW, Petrov DA (2010). Genome-wide patterns of
adaptation to temperate environments associated with transposable elements in
Drosophila. PLoS Genet 6: e1000905.
González J, Macpherson JM, Petrov DA (2009). A recent adaptive transposable element
insertion near highly conserved developmental loci in Drosophila melanogaster.
Molecular Biology and Evolution 26: 1949–1961.
Harada K, Yukuhiro K, Mukai T (1990). Transposition Rates of Movable Genetic Elements
in Drosophila-Melanogaster. Proc Natl Acad Sci USA 87: 3248–3252.
Hedrick PW, Ginevan ME, Ewing EP (1976). Genetic Polymorphism in Heterogenous.
Annual Review of Ecology and Systematics: 1–32.
Hillel J, Simchen G, Feldman MW (1973). Mating Systems and Population Structure in 2
Closely Related Species of Wheat Group .2. Environmental Factors and Population
Structure. Heredity (Edinb) 30: 73–83.
Holsinger KE (1986). Dispersal and Plant Mating Systems - the Evolution of Self-
Fertilization in Subdivided Populations. Evolution 40: 405–413.
Houle D, Nuzhdin SV (2004). Mutation accumulation and the effect of copia insertions in
Drosophila melanogaster. Genet Res 83: 7–18.
Hoyle RB, Ezard THG (2012). The benefits of maternal effects in novel and in stable
environments. J R Soc Interface 9: 2403–2413.
Hu TT, Pattyn P, Bakker EG, Cao J, Cheng J-F, Clark RM, et al. (2011). The Arabidopsis
lyrata genome sequence and the basis of rapid genome size change. Nature Publishing
Group 43: 476–481.
Hudson RR, Kreitman M, Aguade M (1987). A Test of Neutral Molecular Evolution Based
on Nucleotide Data. Genetics 116: 153–159.
Innan H, Kim Y (2004). Pattern of polymorphism after strong artificial selection in a
domestication event. Proc Natl Acad Sci USA 101: 10667–10672.
Jain SK, Marshall DR (1967). Population Studies in Predominantly Self-Pollinating Species
.X. Variation in Natural Populations of Avena Fatua and a Barbata. American Naturalist
101: 19–&.
Jump AS, Hunt JM, Martínez-Izquierdo JA, Peñuelas J (2006). Natural selection and climate
change: temperature-linked spatial and temporal trends in gene frequency in Fagus
sylvatica. Molecular Ecology 15: 3469–3480.
Kamran-Disfani A, Agrawal AF (2014). Selfing, adaptation and background selection in
finite populations. Journal of Evolutionary Biology 27: 1360–1371.
Kazazian HH (2004). Mobile Elements: Drivers of Genome Evolution. Science 303: 1626–
17
1632.
Kidwell MG, Lisch D (1997). Transposable elements as sources of variation in animals and
plants. Proc Natl Acad Sci USA 94: 7704–7711.
King D, Snider LD, Lingrel JB (1986). Polymorphism in an androgen-regulated mouse gene
is the result of the insertion of a B1 repetitive element into the transcription unit. Mol
Cell Biol 6: 209–217.
Kirkpatrick M, Lande R (1989). JSTOR: Evolution, Vol. 43, No. 3 (May, 1989), pp. 485-503.
Evolution.
Kloeckener-Gruissem B, Freeling M (1995). Transposon-induced promoter scrambling: a
mechanism for the evolution of new alleles. Proc Natl Acad Sci USA 92: 1836–1840.
Kozak GM, Boughman JW (2012). Plastic responses to parents and predators lead to
divergent shoaling behaviour in sticklebacks. Journal of Evolutionary Biology 25: 759–
769.
Lacey EP, Herr D (2000). Parental effects in Plantago lanceolata L. III. Measuring parental
temperature effects in the field. Evolution 54: 1207–1217.
Langley CH, Montgomery E, Hudson R, Kaplan N, Charlesworth B (1988). On the role of
unequal exchange in the containment of transposable element copy number. Genet Res
52: 223–235.
Levene H (1953). Genetic Equilibrium When More Than One Ecological Niche is Available
on JSTOR. American Naturalist 87: 331–333.
Levin D, Kerster H (1974). Gene Flow in Seed Plants (T Dobzhansky, M Hecht, and W
Steere, Eds.). : 139–220.
Li J, Jiang H, Wong W (2010). Modeling non-uniformity in short-read rates in RNA-Seq
data. Genome Biol 11: R50.
Lister C, Jackson D, Martin C (1993). Transposon-induced inversion in Antirrhinum
modifies nivea gene expression to give a novel flower color pattern under the control of
cycloidearadialis. The Plant Cell 5: 1541–1553.
Lohe AR, Moriyama EN, Lidholm DA, Hartl DL (1995). Horizontal Transmission, Vertical
Inactivation, and Stochastic Loss of Mariner-Like Transposable Elements. Molecular
Biology and Evolution 12: 62–72.
Long JC, Knowler WC, Hanson RL, Robin RW, Urbanek M, Moore E, et al. (1998).
Evidence for genetic linkage to alcohol dependence on chromosomes 4 and 11 from an
autosome-wide scan in an American Indian population. Am J Med Genet 81: 216–221.
Loveless MD, Hamrick JL (1984). JSTOR: Annual Review of Ecology and Systematics, Vol.
15 (1984), pp. 65-95. Annual Review of Ecology and Systematics.
18
Lynch M, Walsh B (1998). Genetics and analysis of quantitative traits.
Mackay T (1989). Transposable Elements and Fitness in Drosophila-Melanogaster. Genome
31: 284–295.
Mackay T, Lyman RF, Jackson MS (1992). Effects of P-Element Insertions on Quantitative
Traits in Drosophila-Melanogaster. Genetics 130: 315–332.
Marshall DJ (2008). Transgenerational plasticity in the sea: context-dependent maternal
effects across the life history. 89: 418–427.
McClintock B (1950). The origin and behavior of mutable loci in maize. Proc Natl Acad Sci
USA 36: 344–355.
McClintock B (1956). Controlling Elements and the Gene. Cold Spring Harbor Symposia on
Quantitative Biology 21: 197–216.
Mitchell-Olds T, Willis JH, Goldstein DB (2007). Which evolutionary processes influence
natural genetic variation for phenotypic traits? Nat Rev Genet 8: 845–856.
Montgomery E, Charlesworth B, Langley CH (1987). A Test for the Role of Natural-
Selection in the Stabilization of Transposable Element Copy Number in a Population of
Drosophila-Melanogaster. Genet Res 49: 31–41.
Mousseau TA, Uller T, Wapstra E, Badyaev AV (2009). Evolution of maternal effects: past
and present. Philosophical Transactions of the Royal Society B: Biological Sciences 364:
1035–1038.
Narain P (1966). Effect of linkage on homozygosity of a population under mixed selfing and
random mating. Genetics 54: 303–314.
Nordborg M (1997). Structured coalescent processes on different time scales. Genetics 146:
1501–1514.
Nordborg M (1999). The Coalescent with Partial Selfing and Balancing Selection: An
Application of Structured Coalescent Processes on JSTOR. Lecture Notes-Monograph
Series.
Nordborg M, Hu TT, Ishino Y, Jhaveri J, Toomajian C, Zheng H, et al. (2005). The pattern
of polymorphism in Arabidopsis thaliana. Plos Biol 3: e196.
Nuzhdin SV, Mackay T (1995). The Genomic Rate of Transposable Element Movement in
Drosophila-Melanogaster. Molecular Biology and Evolution 12: 180–181.
Nuzhdin SV, Pasyukova EG, Mackay TF (1997). Accumulation of transposable elements in
laboratory lines of Drosophila melanogaster. Genetica 100: 167–175.
Oliver KR, Greene WK (2012). Transposable elements and viruses as factors in adaptation
and evolution: an expansion and strengthening of the TE-Thrust hypothesis. Ecol Evol
19
2: 2912–2933.
Palumbi SR (1994). Genetic Divergence, Reproductive Isolation, and Marine Speciation.
Annu Rev Ecol and Syst 25: 547–572.
Pasyukova EG, Nuzhdin SV, Morozova TV, Mackay TFC (2004). Accumulation of
transposable elements in the genome of Drosophila melanogaster is associated with a
decrease in fitness. J Hered 95: 284–290.
Pimpinelli S, Berloco M, Fanti L, Dimitri P, Bonaccorsi S, Marchetti E, et al. (1995).
Transposable Elements Are Stable Structural Components of Drosophila-Melanogaster
Heterochromatin. Proc Natl Acad Sci USA 92: 3804–3808.
Plaistow SJ, Benton TG (2009). The influence of context-dependent maternal effects on
population dynamics: an experimental test. Trends in Ecology & Evolution 364: 1049–
1058.
Pollak E (1987). On the theory of partially inbreeding finite populations. I. Partial selfing.
Genetics 117: 353–360.
Reed TE, Waples RS, Schindler DE, Hard JJ, Kinnison MT (2010). Phenotypic plasticity and
population viability: the importance of environmental predictability. Proceedings of the
Royal Society B: Biological Sciences 277: 3391–3400.
Reilly MT, Faulkner GJ, Dubnau J, Ponomarev I, Gage FH (2013). The role of transposable
elements in health and diseases of the central nervous system. J Neurosci 33: 17577–
17586.
Roze D, Lenormand T (2005). Self-fertilization and the evolution of recombination.
Genetics 170: 841–857.
Rubin GM, Spradling AC (1982). Genetic-Transformation of Drosophila with Transposable
Element Vectors. Science 218: 348–353.
Sanmiguel P, Bennetzen JL (1998). Evidence that a recent increase in maize genome size was
caused by the massive amplification of intergene retrotransposons. Annals of Botany 82:
37–44.
Scheiner SM (1998). The genetics of phenotypic plasticity. VII. Evolution in a spatially-
structured environment. Journal of Evolutionary Biology 11: 303–320.
Schlichting CD (1986). The evolution of phenotypic plasticity in plants. Annual Review of
Ecology and Systematics 17: 667–693.
Schmidt JM, Good RT, Appleton B, Sherrard J, Raymant GC, Bogwitz MR, et al. (2010).
Copy number variation and transposable elements feature in recent, ongoing adaptation
at the Cyp6g1 locus. PLoS Genet 6: e1000998.
Schmitt J, Gamble SE (1990). The Effect of Distance from the Parental Site on Offspring
20
Performance and Inbreeding Depression in Impatiens capensis: A Test of the Local
Adaptation Hypothesis. Evolution 44: 2022.
Schoen DJ, Morgan MT, Bataillon T (1996). How does self-pollination evolve? Inferences
from floral ecology and molecular genetic variation. Philos Trans R Soc Lond, B, Biol
Sci 351: 1281–1290.
Shapiro JA (2010). Mobile DNA and evolution in the 21st century. Mobile DNA.
Slotte T, Foxe JP, Hazzouri KM, Wright SI (2010). Genome-wide evidence for efficient
positive and purifying selection in Capsella grandiflora, a plant species with a large
effective population size. Molecular Biology and Evolution 27: 1813–1821.
Storm JJ, Lima SL (2010). Mothers Forewarn Offspring about Predators: A
Transgenerational Maternal Effect on Behavior. The American Naturalist 175: 382–390.
Strobeck C (1983). Expected Linkage Disequilibrium for a Neutral Locus Linked to a
Chromosomal Arrangement. Genetics 103: 545–555.
Sultan SE, Bazzaz FA (1993). Phenotypic Plasticity in Polygonum persicaria. II. Norms of
Reaction to Soil Moisture and the Maintenance of Genetic Diversity. Evolution 47:
1032.
Sultan SE, Spencer HG (2002). Metapopulation Structure Favors Plasticity over Local
Adaptation. The American Naturalist 160: 271–283.
Tenaillon MI, Sawkins MC, Long AD, Gaut RL, Doebley JF, Gaut BS (2001). Patterns of
DNA sequence polymorphism along chromosome 1 of maize (Zea mays ssp. mays L.).
Proc Natl Acad Sci USA 98: 9161–9166.
Tien NSH, Sabelis MW, Egas M (2010). The maintenance of genetic variation for
oviposition rate in two-spotted spider mites: inferences from artificial selection.
Evolution 64: 2547–2557.
Turelli M, Barton NH (2004). Polygenic variation maintained by balancing selection:
pleiotropy, sex-dependent allelic effects and G x E interactions. Genetics 166: 1053–
1079.
Via S, Lande R (1985). Genotype-environment interaction and the evolution of phenotypic
plasticity. Evolution: 505–522.
Wright S (1921). Systems of Mating. I. the Biometric Relations between Parent and
Offspring. Genetics 6: 111–123.
Wright S (1943). Isolation by distance. Genetics 28: 114.
Wright SI, Kalisz S, Slotte T (2013). Evolutionary consequences of self-fertilization in plants.
Proceedings of the Royal Society B: Biological Sciences 280: 20130133.
21
Wright SI, Ness RW, Foxe JP, Barrett SCH (2008). Genomic consequences of outcrossing
and selfing in plants. Int J Plant Sci 169: 105–118.
Yamada T, Ohtani S, Sakurai T, Tsuji T, Kunieda T, Yanagisawa M (2006). Reduced
expression of the endothelin receptor type B gene in piebald mice caused by insertion of
a retroposon-like element in intron 1. J Biol Chem 281: 10799–10807.
22
Chapter 2
Genetic variation of transposable element host control
mechanism
The following chapter has been published as a research article in Heredity:
Wendy T. Vu, Sergey V. Nuzhdin: Genetic variation of Copia suppression in Drosophila
melanogaster. Heredity 106.2 (2010): 207-217
2.1 Abstract
Transposable elements (TE) are genomic parasites that persist by exploiting its host
reproductive machinery. However, some hosts within a population have evolved the ability
to silence TE activity while others have not. We are interested in investigating the population
dynamics of TE host silencing pathways, particularly copia long terminal repeat
retrotransposon in Drosophila melanogaster. Here we identify large effect genes involved in copia
suppression by using a semi-quantitative analysis to assay levels of copia plasmids (believed to
be an intermediate of transposition) in ninety-eight recombinant inbred lines constructed
from a line exhibiting high copia transpositions and a line exhibiting no transpositions. The
results revealed that the influence of copia copy number and transcription level on copia
plasmid concentrations are weak and that genomic factors, presumably encoded by the host,
have stronger effects on transposition rates. We mapped a QTL affecting copia plasmid
concentration in the 33A-43E interval and applied a quantitative deficiency
complementation analysis on this chromosomal region. One out of two large effect
deficiencies on copia plasmid concentrations corresponded to a gene called vasa, an important
23
component in the nuage-piRNA TE silencing machinery. We hypothesize that copia
suppression occurs by the joint action of several post-transcriptional mechanisms with at
least one of the blocks taking place in the nuage.
2.2 Introduction
Transposable elements (TE) are selfish genetic elements that propagate through the
population by exploiting the host sexual reproductive machinery. TEs have existed in the
genomes of living organisms for hundreds of millions of years and are found in most plants
and animals; with as much as 90% of the genome being TE derived sequences (Finnegan
1992). For example, there are approximately 100,000 copies of LINE 1 elements and about a
million copies of the Alu elements in humans (see XING et al. (2007); E. S. LANDER et al.
(2001)). As genetic parasites, TEs exploit the host transcriptional system to increase copy
number. Insertions into new chromosomal locations are generally deleterious because they
disrupt gene expression, cause ectopic exchange and chromosomal breakage
(CHARLESWORTH, SNIEGOWSKI and STEPHAN 1994). While some TE-induced mutations are
beneficial, it is clear that their net effect on host fitness is negative, with an estimated fitness
cost in excess of 5% (Charlesworth et al. 1994). How have they continued to parasitize the
genome without self-destructing by means of killing the host? One explanation seems to be
that TE activity resulting from copy number increase is balanced by selection against hosts
with TE induced deleterious effects.
The gypsy element, one of the best-studied retrotransposons, offers a glimpse of genetic
variation in TE transmission and host control factors. Two components are required for the
activity of gypsy: maternally inherited “rogue” or active elements and the permissive flamenco
24
locus (BUCHETON 1995; PRUD’HOMME et al.1995). There are two classes of flamenco alleles,
restrictive and permissive, present in natural populations (PELISSON et al.1997). The
restrictive alleles are dominant and repress the mobility of functional proviruses by
suppressing gypsy transcripts through the piwi RNA-silencing pathway (PELISSON et al. 2007).
The piwi RNA (piRNA) pathway is a maternally transmitted defense mechanism designed to
control the proliferation of a variety of TE families, such as TAHRE retroelement (SHPIZ et
al. 2007) and P transposons (SIMMONS et al. 2007). piRNAs are 29 nucleotide RNA
molecules that interact with silencing proteins Piwi, Aubergine, and Argonaut, which are
involved in transcriptional and chromatin silencing of retrotransposons in the D. melanogaster
germ line (KLENOV et al. 2007). Worth noting is that piRNA has so far been shown to act
through decreasing transcript levels by either direct silencing of transcription or RNA-decay
(GRIVNA et al. 2006; HUTVAGNER et al. 2008).
We focus on characterizing a potential host control mechanism for the copia element. Similar
to the gypsy element, copia is a long-terminal-repeat (LTR) retrotransposon. However, unlike
the TAHRE retroelement and P transposons, copia activity is suppressed in genotypes
exhibiting high copia transcript levels (NUZHDIN et al. 1998). Here we analyze the genetics of
this unusual post-transcriptional suppression mechanism. Copia-Ty1 belongs to the
Psuedoviridae retrotransposon family (PETERSON-BURCH and VOYTAS 2002) and is
abundant in plants, present in insects (1731 and copia families in Drosophila), but uncommon
in vertebrates. Copia transpositions take place primarily in fly spermatocytes, where their
transcript level is strongly elevated in the early stages of spermatogenesis (PASYUKOVA et
al.1997). Copia is stable in the majority of genotypes, including Oregon R (Ore), but active in
several wild-type strains, such as 2b (PASYUKOVA and NUZHDIN 1993). There is a strong
25
positive correlation between copy number and transposition rate, suggesting that copia copy
number is not self-regulated (PASYUKOVA et al.1998). Copia, unlike gypsy, does not require
maternal transmission of a “rogue” copy for transposition to occur (PERDUE and NUZHDIN
2000).
We identified candidate genes and potential control mechanisms that suppress the copia
element in D. melanogaster. According to BOEKE and CHAPMAN 1991 (Figure 2.1), virus-like-
particles (VLP) are assembled, during the late stages of the transposition pathway and serve
as sites for RNA reverse transcription into linear and circular DNA transposition
intermediates. For this reason, we quantified copia extrachromosomal DNA plasmids in 98
recombinant inbred lines (RILs) for QTL analyses.We mapped regions of the genome
required for copia reverse transcription and conducted a deficiency mapping analysis to
localize candidate genes within these QTL regions that interfere with copia reverse
transcription. The most interesting gene represented in the analysis is the vasa gene, which is
expressed in a special compartment of the germ cell called the nuage. In the cytoplasm, the
nuage interacts with the PIWI protein, which is an important component of the piRNA
machinery (KLATTENHOFF and THEURKAUF 2009). Therefore, we propose an unusual TE
suppression mechanism that requires transcript sequestering during critical stages of
spermatogenesis.
26
Figure 2.1 Copia transposition process. a. copia transcription. b. copia mRNA processing and
translation of gag, pol, int and Rnase transcripts in the cytoplasm. c. Translated proteins are delivered
into the nucleus and VLPs are assembles from gag proteins. d. Inside the VLPs, reverse transcription
of full-length mRNA into DNA. e. Complex strain exchanges required to fully replicate the two long
terminal repeats (LTRs) to produce linear DNA, single and double LTR extrachromosomal plasmids.
Finally, integration of a newborn element into the host genome occurs.
2.3 Materials and Methods
2.3.1 Wild-type Drosophila melanogaster lines and copia transpositions
We used the two isogenic lines 2b (PASYUKOVA et al. 1998), Ore (Oregon R, NUZHDIN et al.
1996), and a set of recombinant inbred lines (RILs) derived from them (NUZHDIN et al.
1997). Briefly, in the Ore line, no transpositions have been found over six years and copia
positions are fixed (NUZHDIN et al. 1998). In the 2b line, a high copia transposition rate (10
-3
-
27
10
-2
) was observed in 1991 (PASYUKOVA and NUZHDIN 1992) and has continued since then
(PASYUKOVA et al. 1998). Although copia has fixed in some sites, new transpositions segregate
in the 2b line. We generated a set of 98 recombinant inbred lines by crossing one 2b male
with an Ore female, then backcrossed F1 females with a 2b male, with repeated full-sib
mating of subsequent progeny thereafter (see NUZHDIN et al. 1997 for details). We mapped
copia positions (either fixed or segregating) in two replicate parental 2b males (Table 2.1).
Accordingly, RILs vary not only in genetic composition, but also in the copy number of copia
elements. Furthermore, several copia transpositions were detected during and after the
construction of the RILs (see NUZHDIN et al. 1998 for detail). We used transcript level, copy
number and transposition activity generated from NUZHDIN et al. (1998) in our analysis of
the copia element in each of the RILs. Transcript level was evaluated by standard Northern
analysis. Cytological positions were obtained by in situ hybridization of the plasmid
cDM5002 containing a full-length copia element (FINNEGAN et al. 1978) to polytene salivary
gland chromosomes of third instar larvae (SHRIMPTON et al. 1986). Probes were labeled with
biotinylated dATP (bio-7-dATP, BRL) by nick translation. Hybridization was detected using
the Vectastain ABC kit (Vector Labs) and visualized with diaminobenzidine. The element
locations were determined at the level of cytological bands on the standard Bridge’s map
(LEFEVRE 1976).
2.3.2 Amplification of copia plasmid from parental lines Ore and 2b
During retrotransposition, copia DNA plasmids are produced containing either a single LTR
or double LTR. Although, single LTR plasmids are thought to be abortive, it is unclear
whether double LTRs (dLTR) are abortive or direct intermediates of copia transposition
(FLAVELL AND ISHOROWICS 1981; FLAVELL 1984). According to the copia sequence reported
28
by CSINK AND MCDONALD (1995), We designed primers to amplify the junction between the
5’ and 3’ LTRs of the copia dLTR plasmids (Table 2.1). Six different primer pairs were
designed to represent copia sequence variance (P1/P2, P1/P3, P1/P4, P1/P5, P1/P6,
P1/P7).
DNA samples were prepared from 10 males aged for approximately one week using the
Puregene Kit (Gentra). The DNA concentrations were determined on a Beckman
Specktrophotometer Du-65. PCR amplifications were carried out in the 21 ul reaction
containing 200 ng genomic DNA, 20ng of the primer P1 and one of the primers P2-P7,
1xPCR Buffer, 200 uM dNTP, 3mM magnesium chloride, and 2.5 U of Taq polymerase
(Promega) 0.15 microCi of gama-
33
P (Amersham) per amplification was incorporated into
primer P1 with 0.1 U of T4 Polynucleotide Kinase (New England Biolabs) in a buffer
supplied by the manufacturer. Thermal cycle conditions were: 5 min at 94°C, followed by 35
cycles for 0.5 min at 94°C and 1 min at 60°C, followed by a 5 min final extension at 72°C.
The amplification products were visualized after fractionation through 6% acrylamide gel,
using Phosphorimaging on the Storm 8600 imager.
29
Table
2.1
Deficiency
effects
of
copia
plasmid
concentrations.
P-‐values
correspond
to
the
effects
of
parental
line
(2b
versus
Ore;
chromosome
inherited
from
a
deficiency
stock
-‐
balancer
versus
deficiency;
and
their
interaction
term)
2.3.3 Electron microscopy images of VLPs in testes
Since VLPs are believed to be primary vehicles for retrotransposition (BOEKE and
CHAPMAN 1991), we took electron microscopy images of testes cross-sections to validate this
assumption. Testes were dissected in phosphate buffer (pH = 7.4) and fixed in 2%
paraformaldehyde and 0.1% glutaraldehyde in phosphate buffer (pH 7.4) for 12 h at room
temperature; then placed in 2% osmium tetroxide and cacodylate buffer for storage at 4°C.
The fixed testes were dehydrated through an ethanol (EtOH) dilution series up to 100%
EtOH, infiltrated in a 1:1 EtOH / LR White mixture overnight, embedded in 100% LR
White acrylic resin (Ted Pella Inc., Redding CA) in beam capsules, and incubated overnight
at 60 °C. The blocks were then ultra thin sectioned (75 nm in thickness) and placed on
parlodian coated nickel grids. Sections on grids were etched with 0.5% sodium
Deficiency Lines Break Points Parental Effect
Effect of
Deficiency
Parent Deficiency
Interaction
Df(2R)ED1715 43A4—43F1 0.0052
0.0028 0.6317
Df(2R)ED1673 42E4—43D3 0.9185
0.8802 0.5786
Df(2R)ED1618 42C3—43A1 0.0010
0.0011 0.4306
Df(2R)ED1552 42A11—42C7 0.9360
0.1179 0.6425
Df(2R)ED1484 42A2—42A14 0.1357
0.5090 0.6997
Df(2L)ED791 34E1—35B4 0.0871
0.1676 0.6662
Df(2L)ED784 34A4—34B6 <0.0001
0.0599 0.1244
Df(2L)ED780 33E4—34A7 0.7695
0.6655 0.1095
Df(2L)ED761 33A2—33E5 0.0244
0.0216 0.4038
Df(2L)ED1473 39B4—40A5 0.2975
0.0441 0.9268
Df(2L)ED1451 38F5—39E2 0.0013
0.4718 0.4496
Df(2L)ED1315 38B4—38F5 0.0020
0.5130 0.5000
Df(2L)ED1303 37E5—38C6 0.0023
0.0043 0.6030
Df(2L)ED1272 37C5—38A2 0.0337
0.0191 0.4599
Df(2L)ED1203 36F7—37C5 0.0021
0.1246 0.6248
Df(2L)ED1196 36E6—37B1 <0.0001
<0.0001 0.7848
Df(2L)ED1102 35F12—36A10 0.1917
0.7719 0.2454
Df(2L)ED1054 35B10—35D4 <0.0001
<0.0001 0.006
30
metaperiodate 10 min at room temperature to remove excess resin, and then washed 5 times
for 10 min each by drops of 0.025 M Tris buffer (pH 7.4). Sections were viewed and
photographed on a Philips CM10 electron microscope.
2.3.4 Copia double LTR plasmid sequencing
To verify that we are isolating copia dLTR plasmids, we sequenced the junction between the
LTRs (Figure 1d-e. for dLTR structure). DNA was extracted from individual flies and
primers P8 and P9 were designed to amplify the junction between the 5’ and 3’ LTR of the
dLTR plasmid DNA (Table 2.1). PCR reaction mixture consisted of 1X AmpliTaq Gold
PCR Buffer, 2U of AmpliTaq Gold DNA Polymerase, 100mM MgCl, 40 mM of dNTP’s, 1
mM of each primer and 200 ng of DNA template. Thermo cycling conditions were: 10 min
at 94C, 15 sec at 94 C, 30 sec at 60C, and 15 sec at 72C for 30 cycles. PCR products were
visualized on 1.5% acrylamide gel. We cut a band of the expected size (180 bp) from the gel
and extracted the product from the gel using the Wizard SV Gel and PCR Clean-Up System
(Promega). The extracted product was prepared for sequencing as follows: 6.4 pmol of copia
P9 primer, 40 ng of template and deionized water to bring total volume to 18 ul. The sample
was submitted to USC Norris Core Facility for sequencing.
31
Primers/Probes Sequence
P1 5’-GTCGTGGTGCTGGTGTTGCAGTTG-3’
P2 5’-GAATAAAAAGAGTGGTATTCTCTT-3’
P3 5’-GAATAAAAAGAGTGGTATTCTCTC-3’
P4 5’-CACAGCAAAAAACGTACAAGAAGA-3’
P5 5’-CACAGCAAAAAACGTACAAGAAGG-3’
P6 5’-ACGTACAAGAAGGAAAGAAGGAAA-3’
P7 5’-ACGTACAAGAAGGAAAGAAGGATT-3’
P8 5’-AGGTGTGGCCATTCATATCAAATA-3’
P9 5’-GTGCTGGTGTTGCAGTTGAA-3’
P10 5’-CTCCACCAGGAAACGACATT-3’
P11 5’-TTCCGTTAAGCATTGCCTTC-3’
P12 5’-FAM6- CAGCAACTACGCGCAGAGCT-3’
P13 5’- FAM6- TACAACATGTTGGAATATAC - 3’
Table 2.2 Copia plasmid LTR junction Primers and Probes.
2.3.5 Semi-quantitative analysis of concentration of copia plasmid in each RIL
To get an idea of the relative differences in copia dLTR plasmids between the 2b and Ore
parental lines, we extracted DNA and semi-quantified relative copia plasmid concentrations,
as mentioned above. In order to measure the amount of target dLTR molecules, we
compared the concentration of copia plasmids relative to a known quantity of pBB54 plasmid
containing two tandemly arranged LTRs separated by 15bp insertion 5’-
AGGTGAAAAGGTTTC-3’ (FLAVELL and ISH-HOROWICZ 1983). Both the total DNA and
pBB54 plasmid were amplified in the same reaction mixture that included primers P1 and P4
with 0.15 microCi of gama-
33
P (Amersham) per amplification. Before amplification, the
plasmid was linearized with restriction enzyme Hind III (Promega) at 1 ug per 6,000 ul and 1
ug at 60,000 ul. Each PCR amplification contained 200ng of total DNA, 20 ng of each
primer, and 1 ul of the plasmid pBB54. The intensity of the amplified products were
recorded after fractionation through 6% acrylamide gel by Phosphorimaging on the Storm
8600 imager. The amount of DNA (
33
P) in each band was approximated from the signal
level in the same area rectangles covering the bands. The relative concentrations of the
32
double LTR circles were estimated as a ratio of the signal intensity of the genomic DNA
template band to the plasmid DNA template band generated from the same reaction
mixture. Across 98 RILs and 2 parental lines, the average ratio was 0.472 for the plasmid
concentration 1/6,000 ug/ul, and 4.81 for the plasmid concentration 1/60,000 ug/ul.
Correlations between copia copy numbers, transcript levels and plasmid concentrations
among RILs were calculated with the CORR procedure on data averaged between
observations within each genotype (SAS Institute 1989).
2.3.6 QTL mapping for copia plasmid concentration
Each RIL has been genotyped using 92 roo TE with polymorphic markers on the X, second,
third and forth chromosomes (NUZHDIN et al.1997) to map factors controlling concentration
of copia plasmid. We used the marker data available from NUZHDIN et al. (1997) to map
factors controlling concentration of copia plasmid. We replicated statistical techniques and
directly linked our new analyses to those described by NUZHDIN et al. (1998). Briefly, we
used the QTL Cartographer software (BASTEN, WEIR and ZENG 1997) for composite
interval mapping (ZENG 1994) to test whether an interval flanked by two adjacent markers
actually contains a QTL affecting the trait of interest. With multiple regression analysis of
markers outside of the test interval, we can control for the effects of QTLs that are
chromosomally linked. 76 cytological markers (16 markers out of 92 were completely linked
with neighboring markers and excluded from the analysis, NUZHDIN et al. 1997) were used
for the analysis with parameters 6 (model) and 10 (window size). The conditioning markers
were chosen by stepwise forward regression. The likelihood ratio test statistic, LR, is -
2ln(L
0
/L
1
), where L
0
/L
1
is the ratio of the likelihood under the null hypothesis (there is no
QTL in the interval) to the alternative hypothesis (there is a QTL in the interval). An
33
empirical distribution of LR test statistics under the null hypothesis of no association
between any of the intervals and the trait values were obtained by randomly permuting the
trait data 1000 times and calculating the maximum LR statistics across all intervals for each
permutation. LR statistics from the original data that were exceeded by the permutation
maximum LR statistics less than 50 times are significant at p=0.05. The trait values were the
ratios of the band intensities averaged over 2 (3) measurements per line as described below.
The analysis of the log-transformed data yielded the same results (data not shown).
2.3.7 Deficiency mapping of copia plasmid concentration
We obtained Drosdel deficiency stocks from the Drosophila Genetic Resource Center
(DGRC) in Kyoto, Japan. The deficiencies span the 33E-43E cytological regions on the
second chromosome and maintained with SM6a balancer chromosomes. In choosing a
subset of deficiencies, we attempted to minimize their number while maximizing coverage of
the QTL controlling the amount of copia plasmid as described in the results section. Virgin
females from isogenic 2b and Ore lines were crossed separately with males from each
deficiency line. F
1
progeny consisted of 4 classes: Df/2b, SM6a/2b, Df/Ore, SM6a/Ore (Df
corresponds to a deficiency and SM6a corresponds to a balancer chromosome with a
dominant curly wing marker). We made two replicates of each cross consisting of 5 virgin
females and 3 males in 5 ml vials with standard cornmeal-agar-yeast-corn syrup-malt-soy-
yeast medium and maintained at 25C. F
1
males from each genotype were collected and aged
for 3-5 days before freezing in liquid nitrogen for DNA extractions.
The relative concentration of copia dLTR plasmids was analyzed using DNA samples from F
1
males collected from the 2b and Ore deficiency crosses. Replicate DNA samples were
34
extracted from 5 F1 males per genotype. DNA extractions were carried out in a 96-well
format using ABI Prism 6100 Nucleic Acid Prepstation (Applied Biosystems, Inc.) by
following the manufacturers suggested DNA extraction protocol. The concentrations of
extracted DNA were spectrophotometrically quantified. The Taqman assay of the real-time
PCR detection technique was used to quantify relative concentrations of copia plasmids with
the comparative method for relative quantification (Applied Biosystems, Foster City, CA).
We chose mdg3 as the endogenous reference gene, a LTR retro-transposable element
reported to be a stable element in both parental lines (PASYUKOVA and NUZHDIN 1993).
Amplification efficiencies for both target and reference genes should be between 90-105%
(Real-Time PCR Applications Guide, Bio-Rad). Preliminary experiments consisted of 5 ul of
a 10 fold dilution series between 100-0.05 ng/ul of DNA extracted from 2b parental lines,
targeting both mdg3 and copia plasmids to optimize experimental conditions.
Since Taqman based qPCR assays are very sensitive, the primers used in the semi-
quantitative analysis did not produce efficient amplification of copia plasmids. We designed a
new set of primers maintaining the same conceptual design in this experiment. The Primers
and probes were designed using Primer3 web-based software
(http://fokker.wi.mit.edu/primer3/input.htm). mdg3 primers (P10/P11) and probe (P12)
were chosen to flank the region between the 5’ and 3’ LTR in the genome. Copia primers P8
and P9 are described above and the copia probe (P13) was designed to hybridize to the
complementary strand of the 5’ and 3’LTR junction. PCR amplifications were performed in
96-well reaction plates, using separate wells to detect copia and mdg3 sequences. We also
included multiple replicates of calibrator samples and a 10-fold dilution series of a sample 2b
DNA template for copia and mdg3 sequences to ensure reaction efficiencies are between 90-
35
105% for each plate. The reaction mixture consisted of 12.5 ul of Taqman Universal PCR
Master Mix (Applied Biosystems), 500nM of each primer, 250 nM of each Taqman mdg3
FAM Probe (Applied Biosystems), 30-60 ng of DNA template and deionized water to bring
the total reaction volume to 25 ul. The real-time PCR amplification was performed in an MJ
Thermocycler (Bio-Rad) with the following conditions: 95C, 10 min; 95C, 15 sec; 60C, 1
min; for 50 cycles.
Since the genomic DNA samples contain both chromosomal and extrachromosomal DNA,
the target copia plasmid amplification efficiencies were low. As a result, we verified the
efficiency of the experimental conditions using PCR products amplified with P8/P9 primers.
2b, Ore and deficiency parental genotypes were used as calibrators to compare and estimate
the threshold cycle (Ct) values (i.e. the fractional cycle number at which the amount of
amplified sequence reaches the threshold, Bio-Rad) for the F
1
samples. The amount of target
sequence (copia) was normalized to the reference sequence (mdg3) and compared to the
calibrator samples (parental genotypes: 2b/Ore/Def). The calculations to estimate the
relative expression ratio of the target and calibrator samples are as follows:
ΔCt = (target)copia Ct – (ref)mdg3 Ct
ΔCt =(calibrator)copia Ct – (ref) mdg3 Ct
ΔΔCt = ΔCt(target) - ΔCt(calibrator)
Relative Expression Ratio = 2
-
ΔΔ
Ct
36
2.4 Results
2.4.1 Copia plasmid in parental stocks
In the process of transposition, copia is believed to produce dLTR plasmid products that may
be direct intermediates of retrotransposition (FLAVELL and ISH-HOROWICZ 1981). To test
the parental strains 2b and Ore for the presence of dLTR copia plasmid, we amplified the
junction between the dLTRs. Although Copia elements and retroviruses both share similar
transposition mechanism along with structural and genetic sequence homology, they differ in
the mode of integration and excision. Retroviruses require an additional two base pair
overhang for integration, while copia Ty1 can integrate with blunt ends (BOEKE and
CHAPMAN 1991).
Similar to that seen in the pseudoviridae retrotransposon, we detected an exact junction
without the additional two base pair insertion between the dLTRs for copia plasmids
(FLAVELL 1983). In the 2b and Ore lines, we used four different combinations of primers
and observed the expected size bands of approximately 260bp (P1/P4 and P1/P5), 310bp
(P1/P2 and P1/P3) and 330bp (P1/P6 and P1/P7) in the 2b line in addition to multiple
faint bands of different sizes from the amplification products, which may be products of
neighboring or scrambled copia copies in the genome with LTRs within close proximity
(Supplementary Figure 2.2). We also observed multiple faint bands in the Ore samples, some
partially overlapping with 2b amplified bands. However, there was no apparent amplification
product of the expected size detected in the Ore parental line. These results suggest that
dLTR plasmid structures may be intermediates required for copia transposition.
37
Figure 2.2 Copia plasmids in parental lines. Column 1-6 correspond to copia plasmids isolated from
the active 2b parental line. Bands in column 1-2 reflect 260 bp amplicon products from primers
P1/P4 and P1/P5. Bands in column 3-4 reflect 310 bp amplicon products from P1/P2 and P1/P3.
Bands in column 5-6 reflect 330 bp amplicon products from P1/P6 and P1/P7. Column 7, 9, 11
correspond to 2b parental line amplified with P1/P4, P1/P2, P1/P6 respectively. Column 8, 10, 12
correspond to Ore parental line amplified with P1/P4, P1/P2, P1/P6 respectively.
2.4.2 Electron microscopy images of VLPs in testes
Since VLPs are believed to be the primary vehicle for copia plasmid formation and
transposition (BOEKE AND CHAPMAN, 1991), we took electron micrographs of testes cross-
sections and found distinct differences between 2b and Ore parental lines (Figure 2.3). We
found VLPs in 2b, but not Ore in the serial cross-sections taken from the distal parts of the
testes containing maturing spermatids, specifically after the pre-individualization and before
the coiling stages of maturing spermatids. As described in RACHIDI et al. (2005), we saw
spherical structures composed of clusters of well defined particles approximately 50 nm in
P1/P4! P1/P5 P1/P2
P1/P3 P1/P6 P1/P7 2b3 Ore 2b3 Ore 2b3 Ore
1 2 3 4 5 6 7 8 9 10 11
12
38
diameter, which appear analogous to A-type particles produced in yeast Ty1 LTR
retrotransposons. This might indicate that VLPs and copia plasmids are a necessary part of
the transposition pathway.
Figure 2.3 EM micrographs of testes-cross section parental lines. (A) and (C) shows maturing
spermatid from the 2b and Ore line, respectively, with its axoneme (Ax), major mitochondrial
derivative (M) and minor mitochondrial derivative (m). (B) and (D) corresponds to a higher
magnification of a single spermatid from 2b and Ore, respectively, with virus like particles indicated
by the arrow.
2.4.3 Copia plasmid in RILs
Since plasmids are believed to be copia transposition intermediates, we semi-quantified copia
plasmid concentrations in the 2b and Ore parental lines and found strong differences. We
measured approximate concentrations of copia plasmids in each of the 98 RILs, in order to
map the regions of the 2b and Ore genomes that underlie these phenotypic differences
39
between the parents. The intensity of the 2b parent was ~45 folds stronger than the known
pBB54 plasmid concentration, whereas the Ore parental line and 70% of the RILs displayed
amplification intensities similar to the pBB54 concentration (Figure 2.4). In the remaining
30% of RILs, the amplification intensity of the plasmid product was intermediate between
the two parental strains, but closer to the stable Ore parent. Interestingly enough, it was
previously observed that there was direct copia activity in 5 of the RILs with intermediate
plasmid concentrations (NUZHDIN et al. 1998, Figure 2.4 shown as black bars), suggesting
that these RILs may contain regions affecting copia transposition. Although our assay was
only semi-quantitative, the differences were strong enough to allow us to test our hypothesis.
Figure 2.4 Distribution of copia plasmid concentrations among parental and recombinant inbred
lines. The white and black triangles represent relative copia plasmid concentration of the parental Ore
and 2b line, respectively. Black bars represent RILs exhibiting copia transpositions.
Concentration of copia plasmid
0 1 2 3 4 5
Frequency of genotypes with different concentration
0
20
40
60
80
40
We observed a considerable amount of variation in the RILs, when comparing copia plasmid
concentrations to copy number and transcript levels. There was a moderate positive Pearson
product-moment correlation between copy number and plasmid concentration (r = 0.285, P
= 0.004), shown in Figure 2.5 (filled circles are RILs with direct transpositions).
Interestingly, the copia copy number is moderate in some RILs with high concentrations of
copia plasmids, while some lines with high copy number had low plasmid concentrations.
Similar to the relationship between copia copy number and plasmid concentration, there was
a moderate positive correlation between copia transcript level and plasmid concentration (r=
0.314, P=0.002). Some RILs exhibiting very high transcript levels yielded little to no
detectable plasmids, while some RILs exhibiting relatively low transcript levels had high
plasmid concentrations. These moderate correlations indicate that genomic factors,
presumably encoded by the host, may have additional and possibly stronger effects on copia
activity than copia copies themselves.
41
Figure 2.5 Correlations of plasmid concentration with copy number and transcript level of copia
element. Filled circles are RILs with direct transpositions.
2.4.4 QTL mapping of copia plasmid concentration
We measured copia plasmid concentrations in each of the panel of RILs and mapped QTLs
that account for the genetic variation among these RILs. As QTL mapping outcomes might,
in principle, be sensitive to the procedures of the analyses, we replicated the exact approach
for mapping copia transcript levels among the same set of RILs used in NUZHDIN et al.
(1998). We detected a single QTL in the pericentric region of the second chromosome (33A-
43E) with substantial statistical support with an LR of 39, which is much higher than the
permutation threshold of 24 (Figure 2.6). When we compared our current mapping results
with those from NUZHDIN et al. (1998), the copia plasmid concentration QTL was positioned
precisely within a QTL required for copia transposition. There are at least 2 (possibly 3)
regions of the 2b genome required for copia transpositions (NUZHDIN et al. 1998) but the
Transcript level
0 20 40 60 80 100
Copy number
20 30 40 50 60 70
Concentration of copia plasmid
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
42
copia plasmid concentration mapped only to one region: the pericentric region between the
intervals 33A-43E.
Figure 2.6 Location of QTL for copia plasmid concentration with respect to transcript level and
transposition rate. Plot from double-log likelihood ratio (LR) of composite interval mapping against
recombination distance on the X (A), second (B) and third (C) chromosomes. Horizontal lines,
Bonferroni-corrected LR critical value for experiment-wise α = 0.05.
2.4.5 Deficiency mapping of copia plasmid concentration
We used a quantitative deficiency complementation technique to more precisely localize the
region directly affecting plasmid concentration within the 33A-43E interval on the second
chromosome. We collected 18 deficiency lines covering the candidate pericentrical region
(33A-43E, Figure 2.7) to detect genetic interactions with the parental Ore and 2b alleles that
influence copia plasmid concentrations. Since copia is only active in males and transposition is
not influenced by maternal effects (NUZHDIN et al., 1996), we made separate replicate one
way crosses between males from both parental lines and females from each deficiency line.
DNA was extracted from the F
1
males of 72 genotypes that fall into four classes: 2b/Def,
43
2b/SM6, Ore/Def, and Ore/SM6. Two-way analyses of variance were used to estimate the
significance of the interaction between the parent of origin and the presence of the
deficiency (Table 2.1). In the majority of the genotypes, copia plasmid concentration was
higher in the F
1
progeny with the balancer chromosome as opposed to the deficiency
chromosome.
Figure 2.7 Chromosome position of deficiency. Black solid lines represent deficiency spanning the
pericentrical region of the 2R (region 41-44) and 2L (region 33-40) chromosome.
44
F
1
progeny from the crosses of the two deficiency lines, however, showed significant
deviations from the patterns described above. The copia plasmid concentrations were higher
in the 2b/Def and Ore/Def genotypes compared to the majority of the deficiencies
(P<0.0001) in the Df(2L)1196 line, with breakpoints 36E6 and 37B1. Moreover, the
deficiency containing F1 genotypes (2b/Def and Ore/Def) for both parental lines had a
much higher plasmid concentration than the progeny containing balancer chromosomes
(2b/SM6 and Ore/SM6) of approximately 4 fold changes on the log scale, P<0.0001. We
conclude that the deficiency had a strong effect on the copia reverse transcription process,
irrespective of the 2b versus Ore alleles of the parent (P=0.78).
In contrast, Df(2L)1054 with the breakpoints 35B10 and 35D4, exhibited epistatic
interactions with the parental alleles. In the F
1
progeny from the Ore and Df(2L)1054 cross,
the genotypes containing balancers (Ore/SM6) had higher concentrations of copia plasmid
than the ones containing the deficiency chromosome (Ore/Def). This is much like the
typical pattern observed for all the deficiencies described above. However, the F
1
genotype
with the 2b parental and Df(2L)1054 chromosome (2b/Def) displayed elevated levels of
copia plasmid concentration, which indicate that this region might be involved in copia
plasmid regulation. The main effect of the parent (P<0.0001), deficiency versus balancer
(P<0.0001), and the interaction term between them (P=0.006) were highly significant. This
indicates that there is an interaction between the deficiency and the 2b-originated alleles
defining copia plasmid concentrations. Out of approximately 50 genes represented in the
deficiency, the vasa gene seemed to be the most interesting because it plays a pivotal role in
germ cell development (STYHLER 1998).
45
To test whether vasa directly influences plasmid concentrations, we conducted a
complementation test using a heterozygote loss-of-function vasa mutant maintained over a
CyO balancer (http://flybase.org/reports/FBst0000284.html). Following the same crossing
scheme for the deficiency crosses mentioned in the materials and methods section, we
generated four genotypes: 2b/vas-, 2b/CyO, Ore/vas-, Ore/CyO. As expected, the F
1
genotype of the two possible heterozygote combinations of the 2b chromosome (2b/vas-
and 2b/CyO) displayed high levels of copia plasmids. However, the F1 progeny derived from
the Ore parental line (Ore/vas-) did not show the same patterns observed in the deficiency
crosses. We previously observed lower levels of plasmid concentrations in Ore derived F
1
males containing a deficiency chromosome (Ore/def) than the ones with a balancer
chromosome (Ore/SM6), but the complementation results show that all Ore derived
genotypes (Ore/vas- and Ore/CyO) displayed approximately the same level of plasmids.
Overall, 2b derived genotypes (2b/vas- and 2b/CyO) had higher plasmid concentrations
compared to the Ore derived genotypes (Ore/vas- and Ore/CyO), which is a similar pattern
observed for the Df(2L)1054 deficiency cross. The failure to completely complement may be
due to the fact that Df(2L)1054 and vasa mutant lines carry different balancer chromosomes,
which makes direct comparisons challenging. Another complementation test using a loss-of-
function vasa mutant with an SM6 balancer chromosome comparable to the deficiency
mutants is required to validate that vasa directly influences copia activity. Although the
deficiencies available to us had some gaps in coverage of the QTL support interval, they
were sufficient to identify strong candidate regions that interact with alleles from the
unstable 2b and stable Ore parental lines. At preset we can use these regions to infer possible
genetic interactions that interfere with the general process of copia transposition.
46
2.5 Discussion
In this study we explore the coevolution between copia retrotransposon and the host genome
to understand the potential mechanism that alleviates the detrimental effects caused by
retrotranspositions. The piRNA pathway is thought to serve as a pre-adaptive defense
mechanism against viruses and TEs (BLUMENSTIEL and HARTL 2005; ARAVIN 2005). Novel
TEs spread unsuppressed until the element inserts itself into either a heterochromatic or
euchromatic piRNA generating loci (Brennecke et al. 2007). This results in the production of
piRNA, which then elicits a silencing response by the host that is generally mediated through
direct transcriptional silencing or degradation of TE transcripts (see recent reviews by
PETERS and MEISTER 2007; KLATTENHOFF and THEURKAUF 2007; BUCHON and VAURY
2006; SETO, KINGSTON and LAU 2007; O’DONNELL and BOEKE 2007). However, copia
control seems to deviate from this general mechanism. NUZHDIN et al. (1998) showed that
copia transcript level QTLs did not coincide with the copia transposition QTLs, indicating that
copia transposition is not restricted by transcriptional silencing but instead by a post-
transcriptional mechanism. Since copia transpositions are limited to male spermatocytes, we
propose an alternative piRNA suppression mechanism that involves spermatogenesis-
specific events.
During spermatocyte development in D. melanogaster, transcription ceases during early stages
of spermatogenesis, with limited transcription observed after the first meiosis (see
PARVINEN 2005 for a helpful review). Accordingly, the transcripts required for completion
of spermatogenesis must be produced during the early stages and preserved for use in the
later stages. This kind of translational regulation is mediated by a special membrane-free
47
microtubule-based structure called the nuage, also known as the chromatoid body in
mammals (PARVINEN 2005). The nuage is comprised of a variety of germ cell related
proteins with the most important being the vasa protein, a DEAD-box RNA helicase, which
acts synergistically with the components from the piRNA system Piwi, AGO3, and Argonaut.
The nuage is thought to mediate translational regulation, silencing male specific transcripts
and transcripts originating from transposable elements through complex interactions among
different families of piRNAs (KOTAJA and SACCONE-CORSI 2007; LIM AND KAI 2007).
Therefore, we hypothesize that copia transcripts are suppressed in the nuage by a post-
transcriptional mechanism during spermatogenesis.
Here, we show that the QTL mapping to the 33A-43E region associated with copia
transposition (NUZHDIN et al. 1998) contributes to blocking one or more of the post-
transcriptional stages of copia retrotransposition. The concentration of copia plasmids
purported to be an intermediate for copia transposition (FLAVELL 1984) is strongly increased
in the RILs with 34A-43E pericentrical region originating from the active 2b parental line.
We also establish a strong affect on copia plasmid concentration that is partially accounted for
by the 35B10-35D4 segment (Table 2.1). Among the 54 genes represented within this region,
vasa stands to be a promising candidate because this protein is a critical component of the
nuage-piRNA processing machinery. Vasa is a hallmark for germ cell development and
belongs to a class of proteins that act as RNA chaperones (PARVINEN 2005). Vasa
expression is restricted to testes and ovaries and null mutations of this gene results in
complete sterility (STYHLER 1998). Furthermore, there were strong fertility problems in the
RILs with the pericentrical region inherited from 2b (NUZHDIN et al. 1997). Coincidently,
this commonly observed fertility problem, along with abdomen abnormalities (NUZHDIN
48
AND PASYUKOVA 1991), are among the phenotypes typical for vasa mutants
(http://flybase.org/reports/FBgn003970.html). However, future studies examining the copy
number variation, structure and expression patterns of vasa is necessary to validate the direct
role of vasa in copia activity of the 2b line.
Unfortunately, the deficiency lines for the most potent pericentrical regions on the second
chromosome were not available in the co-isogenic deficiency collection
(http://www.drosdel.org.uk/coverage.php). Accordingly, we were unable to test one
interesting prediction of this hypothesis: the failure to complement 2b deficiency in copia
suppression by one of the piRNA generating loci. At present, our results provide a candidate
post-transcriptional pathway that involves the nuage-piRNA machinery for copia suppression,
but how this suppression is mediated remains unclear.
The only other well-studied element in the pseudoviridae family of retroelements is 1731,
which offers insight as to how the nuage-piRNA machinery might silence TE transcripts.
Much like copia, transcription of 1731 is strongly up regulated in the early stages of
spermatogenesis but 1731 VLPs are not assembled immediately after translation. Rather, the
VLPs are observed in the nuclei during the late stages of spermatogenesis, in which
chromatin condensation would likely make transpositions impossible. RACHIDI et al. (2005)
interpreted these observations by reasoning that, perhaps, 1731 transcripts are silenced or
sequestered in the nuage to alleviate detrimental effects of transposition during critical stages
of germ cell development. Overall, this picture is consistent with the posttranscriptional
blocks of copia transpositions that we have analyzed here.
49
Our experimental results suggest that translational silencing of copia elements through the
nuage-piRNA machinery seems to be a likely hypothesis for host suppression. In our
quantitative study of copia plasmid concentration (NUZHDIN et al. 1996), we detected copia
plasmids in the active 2b but detected little to no copia plasmids in the inactive Ore line.
Consistent with these results, we saw VLPs present in the 2b line but none were observed in
the Ore line in the testes EM images. Both copia and 1731 VLPs are self-assembled in the
nucleus through an autocatalytic process (YOSHIOKA et al. 1990), which serves as a site for
mRNA reverse transcription into linear and plasmid DNA. Therefore, copia silencing is likely
to occur at the post-transcription blocks of spermatogenesis, in which copia transcripts may
be sequestered in the nuage by complex interactions among piRNAs and its associated
proteins.
In the experiments reported here, we have clarified the different stages at which
transpositions of pseudoviridae TEs may be blocked by providing an analogous pathway to
better understand the potential copia control mechanism. We mapped regions of the host
genome controlling copia transpositions, and identified candidate genes and potential
pathways that could explain how copia transposition is controlled during critical stages of
germ cell development. We know that at least some of the mechanisms of TE suppression
are post-reverse-transcriptional because copia transpositions are blocked at multiple stages
with several regions from different chromosomes that interact to suppress or enable copia
mobility.
50
2.6 References
Aravin, AA, Hannon GJ, Brennecke J (2007) The Piwi-piRNA pathway provides an
adaptive defense in the transposon arm race. Science 318: 761-764.
Aravin AA, Klenov MS, Vagin VV, Bantignes F, Cavalli G, Gvozdev VA (2004)
Dissection of a natural RNA silencing process in the Drosophila melanogaster germ
line. Mol. Cell. Biol. 24: 6742-6750.
Basten C, Weir BS, Zeng ZB (1997). QTL Cartographer: A Reference Manual and
Tutorial for QTL Mapping. (Raleigh, N.C.: Department of Statistics, North
Carolina State University), http://statgen.ncsu.edu/ qtlcart/.
Beoke JD, Chapman KB (1991) Retrotransposition mechanisms. Current Opinion Cell. Biol. 3:
502-508.
Blumenstiel JP, Hartl DL (2005). Evidence for maternally transmitted small interfering
RNA in the repression of transposition in Drosophila virilis. PNAS 102: 15965-
15970.
Bucheton A (1995) The relationship between the flamenco gene and gypsy in Drosophila:
how to tame a retrovirus. Trends. Genet. 11: 349-353.
Buchon N, Vaury C (2006) RNAi: a defensive RNA-silencing against viruses
and transposable elements. Heredity 96: 195–202.
Charlesworth B, Sniegowski P, Stephan W (1994) The evolutionary dynamics of
repetitive DNA in eukaryotes. Nature 371: 215-220.
Csink AK, McDonald JF (1995). Analysis of copia sequence variation within and
between Drosophila species. Mol Bio and Evo 12: 83-93.
Eggleston WB, Johnson-Schlitz DM, Engels WR (1988). P-M hybrid dysgenesis does not
mobilize other transposable element families in Drosophila melanogaster. Nature 331:
368-370.
Fennegan DJ (1992). Transposable elements. Curr Opin Genet Dev. 2: 861-7.
Flavell AJ (1984) Role of reverse transcription in the generation of extrachromosomal
copia mobile genetic elements. Nature 310: 514-516.
Flavell AJ, Ishorowicz D (1981). Extrachromosomal circular copies of the eukaryotic
transposable element copia in cultured Drosophila cells. Nature 292: 591-595.
Flavell AJ, Ish-Horowicz D (1983). The origin of extrachromosomal circular copia
elements. Cell 34: 415-419.
51
Grivna ST, Ergin Beyret, Wang Z, Lin H (2006). A novel class of small RNAs in mouse
spermatogenic cells. Genes & Dev 20: 1709-1714.
Haoudi A, Kim MH, Champion S, Best-Balpomme M, Maisonhaute C (1995). The gag
polypeptides of the Drosophila 1731 retrotransposon are associated to virus-like
particles and to nuclei. FEBS Letters 377: 67-72.
Harada K, Yukuhiro K, Mukai T (1990) Transposition rates of movable genetic elements
in Drosophila melanogaster. Proc. Natl. Acad. Sci. U.S.A. 87: 3248-3252.
Hutvágner G, Zamore PD (2002). A microRNA in a Multiple-Turnover RNAi Enzyme
Complex. Science 297: 2056 – 2060.
Ingram VM (1963) The hemoglobins in genetics and evolution. Colombia University Press, New
York.
Klattenhoff C, Theurkauf
W (2008). Biogenesis and germline functions of
piRNAsDevelopment 135: 3-9.
Klenov MS, Lavrov SA, Stolyarenko AD, Ryazansky SS, Aravin AA, Tuschl T, Gvozdev
VA (2007). Repeat-associated siRNAs cause chromatin silencing of
retrotransposons in the Drosophila melanogaster germline. Nucleic Acids Research
35: 5430-5438.
Kotaja N, Sassone-Corsi P (2007) The chromatoid body: a germ-cell-specific RNA-
processing center. Nature Reviews 8: 85-90.
Lefevre G et.al. (1976). A photographic representation and interpretation of the polytene
chromosomes of Drosophila melanogaster salivary glands. in The genetics and biology
of Drosophila. Academic Press, London 31–66.
Lim AK, Toshie K (2007) Unique germ-line organelle, nuage, functions to repress selfish
genetic elements in Drosophila melanogaster. Proc. Natl. Acad. Sci. U. S. A. 104: 6714-
6719.
Marzo M, Puig M, Ruiz A (2008) The Foldback-like element Galileo belongs to the P
superfamily of DNA transposons and is widespread within the Drosophila genus.
Proc. Natl. Acad. Sci. U.S.A. 105: 2957-2962.
Mevel-Ninio M, Pelisson A, Kinder J, Campos AR, Bucheton A (2007) The flamenco locus
controls the gypsy and ZAM retroviruses and is required for Drosophila
oogenesis. Genetics 175: 1615-1624.
Nurminsky DI, Aguiar DD, Bustamante CD, Hartl DL (2001) Chromosomal effects of
rapid gene evolution in D. melanogaster. Science 291: 128-130.
52
Nurminsky DI, Shevelyov YY, Nuzhdin SV, Gvozdev VA (2004) Structure, molecular
evolution and maintenance of copy number of extended repeated structures in the
X-heterochromatin of Drosophila melanogaster. Chromosoma 103: 277-285.
Nuzhdin SV (1999) Sure facts, speculations, and open questions about the evolution of
transposable element copy number. Genetica 107: 129-137.
Nuzhdin SV, Mackay TFC (1994) Direct determination of retrotransposon transposition
rates in Drosophila melanogaster. Genet. Res. 63: 139-144.
Nuzhdin SV, Mackay TFC (1995) The genomic rate of transposable element movement
in Drosophila melanogaster. Mol. Biol. Evol. 12:180-181.
Nuzhdin SV, Pasyukova EG (1991) New approach to polygene mapping: Location of
genes controlling fitness level in related Drosophila melanogaster stocks. Genetika (Russ)
27: 849-859.
Nuzhdin SV, Pasyukova EG, Mackay TFC (1996) Positive association between copia
transposition rate and copy number in Drosophila melanogaster. Proc. Royal Soc. London B
263: 823-831.
Nuzhdin SV, Pasyukova EG, Mackay TFC (1997). Accumulation of transposable
elements in laboratory lines of Drosophila melanogaster. Genetica 100: 167–175.
Nuzhdin SV, Pasyukova EG, Morozova EA, Flavell AJ (1998) Quantitative genetic analysis
of copia retrotransposon activity in inbred Drosophila melanogaster lines. Genetics
150: 755-766.
O'Donnell DA, Boeke JD (2007). Mighty Piwis Defend the Germline against Genome
Intruders. Cell 129: 37-44.
Olsen JC, Swanstrom R (1985). A new pathway in the generation of defective retroviral
DNA. J. Virology 56: 779-789.
Pasyukova E, Nuzhdin SV, Li W, Flavell AJ (1997). Germ line transposition of the copia
retrotransposon in Drosophila melanogaster is restricted to males by tissue- specific
control of copia RNA levels. Mol. Gen. Genet. 255: 115-124.
Pasyukova EG, Nuzhdin S (1993). Doc and copia instability in an isogenic Drosophila
melanogaster stock. Mol. Gen. Genet. 240:302-306.
Pasyukova EG, Nuzhdin SV, Filatov (1998).
The relationship between the rate of
transposition and transposable element copy number for copia and Doc
retrotransposons of Drosophila melanogaster.
.
Genetical Research 72:1-11.
Parvinen M (2005). The chromatoid body in spermatogenesis. Int Journal of Andrology
28:189–201.
53
Pelisson A, Sarot E, Payen-Groschene G, Bucheton A (2007). A novel repeat-associated
small interfering RNA-mediated silencing pathway down regulates
complementary-sense gypsy transcripts in somatic cells of the Drosophila ovary.
J. Virol. 81:1951-1960.
Pelisson A, Teysset L, Chalvet F, Kim A, Prud’homme N et al. (1997). About the origin
of retroviruses and the coevolution of the gypsy retrovirus with the Drosophila flamenco
host gene. Genetica 100: 29-37.
Perdue S, Sergey V. Nuzhdin (2000).
Master Copy Is Not Responsible for the High Rate
of copia Transposition in Drosophila
.
Mol Biol and Evo 17:984-986.
Pesole G, Gissi C, Lanave A, Saccone S (1995). Glutamine synthetase gene evolution in
bacteria. Mol Bio and Evo 12:189-197.
Peters L, Meister G (2007). Argonaute Proteins: Mediators of RNA Silencing
Mol Cell 26: 611-623.
Peterson-Burch BD, Voytas DF (2002). Genes of the pseudoviridae (Ty1/copia)
retrotransposons. Mol. Biol. Evol. 19: 1832-1845.
Prud’homme N, Gans M, Masson M, Terzian C, Butehton A (1995). Flamenco, a gene
controlling the gypsy retrovirus of Drosophila melanogaster. Genetics 139: 697- 711.
Rachidi M, Lopes C, Benichou JC, Hellio R, Maisonhaute C (2005). Virus-like particle
formation in Drosophila melanogaster germ cells suggests a complex translational
regulation of the retrotransposon cycle and new mechanisms inhibiting
transposition. Cyto. Gen. Res. 111: 88-95.
Seto AG, Kingston RE, Lau NC (2007). The Coming of Age for Piwi Proteins.
Molecular Cell 26: 611-623.
Shrimpton A, Montgomery EA, Langley CH (1986). OM mutations in drosophila
ananasae are linked to insertions of transposable elements. Genetics 114: 125-135.
Simmons MJ, Ryzek DF, Lamour C, Goodman JW, Kummer NE, Merriman PJ (2007).
Cytotype regulation by telomeric P elements in Drosophila melanogaster: evidence for
involvement of an RNA interference gene. Genetics 176: 1945-1955.
Shpiz S, Kwon D, Uneva A, Kim M, Klenove M, Rozovsky Y, Georgiev P, Savitsky M,
Kalmykova A (2007). Characterization of Drosophila telomeric retroelement
TAHRE: transcription, transpositions, and RNAi-based regulation of expression.
Mol. Biol. Evol. 24: 2535-2545.
Styhler S, Nakamura A, Swan A, Suter B, Lasko P (1998). vasa is required for GURKEN
accumulation in the oocyte, and is involved in oocyte differentiation and germline
cyst development. Development 125: 1569-1578.
54
Xing JC, Witherspoon DJ, Ray DA, Batzer MA, Jorde LB (2007). Mobile DNA elements
in primates and human evolution. Amer Journal of Phys. Anthro. 134: 2-19.
Xiong Y, Eickbush THA (1990). Origin and evolution of retroelements based upon their
reverse transcriptase sequences. EMBO J. 9: 3353-3362.
Yoshioka K, Fujita A, Kondo S, Miyake T, Sakaki Y, Shiba T (1992). Production of a
unique multi-lamella structure in the nuclei of yeast expressing Drosophila copia
gag precursor. FEBS Letters 302: 5-7.
Yoshioka K, Honma H, Zushi M, Kondo S, Togashi S, Miyake T, Shiba T (1990). Virus-
like particle formation of Drosophila copia through autocatalytic processing. EMBO
Journal 9: 535-541.
Zeng ZB (1994). Precision Mapping of Quantitative Trait Loci. Genetics 136: 1457-1468.
55
Chapter 3
Maintenance of genetic variation and microgeographical
adaptation
This chapter is currently in preparation for submission as a Research Article to Plos Biology.
Wendy T. Vu, Peter L. Chang, Eric von Wettberg, Maren L. Friesen, Tevfik H. Kitapci,
Sergey V. Nuzhdin: Patterns of microgeographical adaptation to a patchy saline
environment in Medicago truncatula.
3.1 Abstract
Divergent selective pressures across a heterogeneous environment can result in the
maintenance of genetic variation and adaptation to local environmental conditions. While
this process has been well documented at larger geographical scales, fewer examples have
been well documented of microenvironmental variation resulting in the maintenance of
genetic variation with none documenting this on a whole genome scale. Saline environments
offer an opportunity to study micro-environmental adaptation due to the high variability in
soil salinity concentrations. Our study examines the patterns of micro-scale adaptation to a
saline habitat within a single selfing population of Medicago trucatula—an annual legume that
is native to Mediterranean regions. Adaptation to microenvironmental variation can result in
either locally specialized ecotypes (phenotypically fixed) or generalist ecotypes
(phenotypically plastic). We investigate this by testing whether adaptive responses vary
predictably along microenvironments or exhibit genotype-environment interactions.
56
Annual plants adapted to saline environments typically flower early to avoid rising salinity
levels as the season progresses. Early flowering confers increased plant performance and
reproductive output in saline conditions. We did not find genotype-environment interactions
in flowering and performance traits, but flowering time is phenotypically fixed and varies
predictably along the salinity gradient. This suggests that variable selective pressure between
soil patches is likely driving functional divergence in flowering time within a single selfing
population. Thus the variability in soil patches is likely a mechanism maintaining phenotypic
variation and potentially genetic differences among ecotypes. Genome resequencing
identified polymorphisms strongly associated with soil salinity levels as well as latitude,
longitude and elevation—all within a single population. Many of these polymorphisms cause
functional changes in genes associated with flowering time, disease resistance, salt and
osmotic stress. Overall, these results suggest that micro-scale environmental variation can
shape patterns of adaptation and maintain genetic variation through differential selection
between micro-soil environments.
3.2 Introduction
The soil environment poses a strong selective pressure on plant fitness and genetic diversity
and plays a critical role in speciation and the maintenance of genetic variation (Levene, 1953;
Mills and Bever, 1998)(Gillespie 1974, 1975; Prout 1968; Ewing 1979). Typically, adaptation
is considered local when populations exhibit higher fitness in their native habitat relative to
foreign environments (Kawecki & Ebert, 2004), where ecotypes exhibit maladaptive
plasticity in foreign environments relative to their home environments. Another indicator of
local adaptation is gene by environment correlations along an environmental gradient
57
(Turesson, 1922; Conover et al., 2009), where adaptive phenotypes are locally fixed and
change predictably across spatial gradients (Olsson and Ågren, 2002; Stinchcombe et al.,
2004). For instance, locally adapted populations exhibit flowering time and bud size variation
that correlates with latitudinal clines in Arabidopsis thaliana (Stinchcombe et al., 2004) and
Lythrum salicaria (Olsson and Ågren, 2002). These studies typically examine adaptation to
large-scale environmental heterogeneity, where geographical distance limits migration and
prevents gene flow that facilitates phenotypic and genetic differentiation.
Although isolation by geographical distance is a strong predictor of population
differentiation (Wright, 1943; Nei, 1972; Loveless and Hamrick, 1984; Palumbi, 1994;
Rousset, 1997), experiments in several plant species such as Impatiens capensis and pallida,
Anthoxanthum odoratum L. and Liatris cylindracea, have demonstrated that geographical
isolation is not essential for population differentiation to occur and that adaptation can occur
on smaller spatial scales (Bradshaw et al., 1965; Antonovics and Bradshaw, 1968; Snaydon,
1970)( Argyres & Schmitt 1991; Huber et al. 2004; von Wettberg et al. 2005). Strongly
variable patterns of selection, as well as segregating quantitative genetic variation, have been
shown on small geographical scales, where water, soil nutrient and light contribute to locally
varying selection (Argyres & Schmitt 1991; Huber et al. 2004; von Wettberg et al. 2005;
Schaal & Levin, 1978; Stratton & Bennington, 1998). A classic study of several grass species
demonstrated that plants growing alongside zinc-coated fences exhibited high tolerance to
zinc, while plants located just a few feet away from the fence were less tolerant (Bradshaw et
al., 1965; AL-HIYALY et al., 1988). Interestingly, a study in a mixed mating population of A.
odoratum distributed along the boundary of metal mines and normal pastures observed
population differences in metal tolerance over short distances along with higher self-
58
fertilization rates than those adapted to normal soils (Antonovics and Bradshaw, 1968;
Antonovics, 1968). Because self-fertility acts as a barrier to gene flow, this lead to gene flow
reduction. The author hypothesized that selection may favor self-fertilization to prevent the
dilution of locally adaptive phenotypes through gene flow from closely adjacent populations.
This study investigates adaptation to microenvironmental variation in a self-fertilizing annual
legume, Medicago truncatula. Like outcrossers, selfers produce gametes through meiosis, but
fertilization only occurs within a single hermaphroditic individual. Selfing rate for M.
truncatula is estimated at 95% (Chaulet and Prosperi, 1994; Bonnin et al., 2001; Siol et al.,
2008). High selfing rates are expected to reduce genetic diversity due to inbreeding (Loveless
and Hamrick, 1984; Charlesworth et al., 1992; Nordborg, 1997; Wright et al., 2013). However,
despite the high selfing rate in M. truncatula, higher than expected levels of genetic variability
have been observed within a single population (Bonnin et al., 2001). Here, we hypothesize
that microenvironmental variation is a mechanism mediating the maintenance of genetic
variation in traits related to fitness and survival in a single selfing population.
Two evolutionary conditions are required to facilitate local microsite adaptation. First, gene
flow between microsites through pollination or seed dispersal must be low enough for the
offspring environment to resemble the parental environment (Slatkin, 1985; Via and Lande,
1985; Holsinger, 1986). For M. truncatula seeds, the estimated amount of seed dispersal is low
between microsites within a single population (Bonnin et al., 2001). Selfers are predicted to
be less successful migrants than outcrossers (Holsinger, 1986) and parent-offspring
correlation is, therefore, a necessary condition for adaptive evolution. A study in a Impatiens
capensis demonstrated that self-fertilized offspring express higher overall fitness when grown
59
within close proximity to the parental plant and significantly lower fitness only 12 meters
away from the parental site (Schmitt and Gamble, 1990). Spatial autocorrelation of microsite
conditions in Impatiens populations likely contributes to this pattern of fine-scale local
adaptation (Huber et al 2004, von Wettberg et al 2005).
Second, phenotypic plasticity alone cannot achieve the optimal phenotype in all
microenvironments (Via and Lande, 1985). Phenotypic plasticity is defined as an individual
expressing alternative phenotypes in response to different environmental conditions (Gause,
1947; Bradshaw, 1965), and this can lead to a single individual expressing adaptive
phenotypic responses to a broad range of environments. In order for adaptive plasticity to
evolve there must be genotypic differences in phenotypic plasticity—this type of genetic
variation is known as genotype-environment interactions (Falconer and Mackay, 1996).
However, in populations with low genetic diversity, as seen in selfing populations, genetic
limitations may constrain the evolution of genotype-environment interactions (Via and
Lande, 1985). A previous reciprocal transplant experiment of four locally adapted M.
truncatula populations demonstrated a lack of genotype-environment interaction, where
phenotypic differences among saline and non-saline populations were phenotypically fixed in
performance and survival related traits (Moriuchi et al., 2014; Friesen et al., 2015), suggesting
that phenotypic plasticity may not have the capacity to evolve or selectively unfavored in
these populations.
Here we test adaption to microenvironmental variation in a single saline habitat located
along the northern coast of Tunisia. Microtopographical variation and distance from the sea
can lead to microsite salinity differences on a scale of tens of centimeters in habitats with
60
saline ground water (Wadleigh and Fireman, 1948; Hajrasuliha et al., 1980; Miyamoto et al.,
2005). In northern Tunisia, M. truncatula grows in Sebkha marshes characterized by heavy clay
soil and dominated by deep rooted salt tolerant shrubs as well as annual grasses and forbs
that flourish during the wet winter and spring when salinity levels are lower, but die back in
the summer when salinity levels rise. In annual plants, flowering time is a critical determinant
of reproductive success (Ollerton and Lack, 1998; Griffith and Watson, 2005; Hall and
Willis, 2006). The timing of flowering has been linked to locally adaptive responses to saline
environments in M. truncatula populations with strong selection for earlier flowering
(Moriuchi et al., 2014; Friesen et al., 2015).
This study describes the patterns of local adaptation to microenvironmental variation in a
single selfing population of M. truncatula originating from a saline habitat exhibiting spatial
and temporal variability in soil salinity levels. We test for genotype-environment interactions
in flowering time and performance related traits to identify whether adaptation to micro-
scale salinity variation is mediated by plastic or fixed (specialized) responses. Consistent with
previous findings, we find a lack of genetic variation in plasticity of flowering time and
performance related traits. Because flowering is phenotypically fixed within this population,
we test whether flowering time is correlated with soil salinity in the field and whether there is
a genetic basis for differences in flowering time along a salinity gradient. Our results show
that flowering time responses vary predictably with soil salinity variability in the field along
with genetic differences associated flowering time differences. Finally, we explore
environmental factors (ie. latitude, longitude and elevation) that influence soil salinity
variation in the field and test whether microenvironmental variation is correlated with allele
frequency within a single population. We find that topological variation explains salinity
61
variability in the field and identified functional polymorphisms significantly associated with
microenvironmental variability.
3.3 Results
In a previous study, we identified patterns of local adaptation in four M. truncatula
populations originating from saline and non-saline soil habitats occurring over 50 km from
each other. We demonstrated that earlier flowering is phenotypically fixed and selectively
favored in saline conditions and identified polymorphisms assorting by soil type (Friesen et
al., 2015). Among the soil assorting polymorphisms, we found non-synonymous SNPs
segregating in saline populations that are associated with flowering time response (Friesen et
al., 2015). These results indicate a genetic basis for adaptive flowering time responses in
geographically isolated populations. In our current study, we sampled 252 accessions from a
larger area around one of these saline sites along the northern coast of Tunisia to test micro-
scale adaptation of a single M. truncatula population exhibiting micro-scale variation in soil
salinity levels.
3.3.1 Absence of genotype-environment interactions: flowering time is fixed and
performance traits are plastic
It has been hypothesized that genotype-environment interactions are necessary for the
evolution of adaptive plasticity in heterogeneous environments (Schmalhausen, 1949;
Gillespie and Turelli, 1989). The question still stands on whether natural selection operating
in spatially and temporally varying environments can result in a single genotype that is fit for
all microenvironments (Via and Lande, 1985; Gillespie and Turelli, 1989; Sultan and Spencer,
2002; Bourne et al., 2014). To understand the genetic and phenotypic response patterns of
62
flowering and performance traits, we conducted a greenhouse experiment of field-collected
seeds grown in 0mM and 100mM salt conditions. Leaf size, number of leaves, branches and
pods were considered as measures of performance. We find no significant genotype-
environment interaction in flowering time and performance traits (Table 3.1), indicating the
absence of genetic variation in phenotypic plasticity. There is significant genetic variability in
flowering time (F
(249,786)
= 3.00, P <0.00001) and no significant treatment effects (F
(1,786)
=
1.41, P =0.23), indicating that flowering time differ between ecotypes but is phenotypically
fixed. In contrast, performance traits show significant treatment effects (Table 3.1): saline
treated plants had significantly smaller leaves (F
(1,745)
= 45.35, P < 0.00001), produced fewer
leaves (F
(1,988)
= 23.30, P <0.0001), fewer branches (F
(1,990)
= 38.31, P < 0.00001) and fewer
pods (F
1,992
= 152.61, P < 0.0001) than non-saline treated plants. Furthermore, saline
treatments have a bigger impact on reproductive output than vegetative growth: salt treated
plants produced 64% fewer pods, 12% smaller leaves, 14% fewer leaves and 20% fewer
branches. In summary, all traits measured demonstrate significant genetic variability;
flowering time is phenotypically fixed, while performance traits exhibit phenotypic plasticity
with no genetic variability in plastic responses.
Source Leaf size Number of
leaves
Number of
branches
sqrt(Flowering
time)
sqrt(Number
of pods)
Line
1.69 *** 2.17*** 38.31*** 3.00*** 2.22***
Treat
45.35*** 23.30*** 1.86*** 1.41 152.61***
Line x Treat
0.933 0.99 0.96 0.98 1.15t
Seed mass
63.20*** 126.03*** 97.49*** 9.48** 25.79***
t 0.10 < P < 0.05, P < 0.05* ; P < 0.001 ** ; P < 0.0001 ***
Table 3.1 Analysis of variance on vegetative and reproductive traits. F-values are reported for line,
treatment, and seed mass for each trait. N=252. Significance is indicated by asterisks.
Interestingly, flowering time and seed size are significantly correlated with growth traits in
early and vegetative development. Plants with higher performance tend to flower earlier,
63
while plants originating from bigger seeds tend to perform better irrespective of greenhouse
treatment conditions (Table 3.2). We included seed size as a covariate in the analysis of
variance (Table 3.1). Seed size had a significant effect on leaf size (F
(1,745)
= 63.2, P <
0.00001), number of leaves produced (F
(1,988)
= 119.63, P <0.00001) and branches (F
(1,990)
=
97.44, P < 0.000001), flowering time (F
(1,786)
=8.80, P < 0.01, Table S1) and pod production
(F
(1,991)
= 25.03, P < 0.00001).
Seed mass
(r)
C.I.
(lower limit, upper limit)
Flowering time
(r)
C.I.
(lower limit, upper limit)
First leaf size 0.4229***
(0.58, 0.64) -0.3685*** (-0.41,-0.31)
Leaf size 0.3638***
(0.29,0.38) -0.3603*** (-0.35,-0.25)
Number of trifoliates 0.4953***
(0.45,0.53) -0.3938*** (-0.43,-0.34)
Number of branches 0.4011***
(0.35,0.44) -0.4044*** (-0.29,-0.17)
Pod Number
0.1114* (0.01,0.11) -0.135*** (-0.18,-0.08)
Seed mass ----
---- -0.1400*** (-0.18,0.08)
t 0.10 < P < 0.05, P < 0.05* ; P < 0.001 ** ; P < 0.0001 ***
Table 3.2 Pearson correlation coefficients (r) for early and vegetative growth traits with respect to
seed mass and flowering time. N = 252; Two-sided 95% confidence intervals (C.I.); level of
significance alpha = 0.05.
3.3.2 Topographical variation influences soil salinity variability
Microtopographical differences as well as proximity to the coast and distance from saline
ground water can all influence the accumulation and distribution of salt levels, creating fine-
scale variability in salinity levels (Walter et al., 2001; Miyamoto et al., 2005; Shi et al., 2005).
Here we explore the factors that may influence salinity variability and test whether these
factors explain the distribution of genetic differences. The field site is located approximately
400 meters south of the Mediterranean Sea. Field soil salinity levels varied within tens of cm
64
across the field site (Figure 3.1) and the distance between collected plants ranged from tens
of cm to 600 meters (Figure. 3.2).
Figure 3.1 Plot of plant positions sampled along the latitudinal and longitudinal axis. Colors
correspond to salinity levels in the field (N=252).
10.447 10.448 10.449 10.450 10.451 10.452
36.709 36.710 36.711 36.712 36.713
Longitude
Latitude
low saline
moderately saline
saline
high saline
36.709 36.710 36.711 36.712 36.713
10.447 10.448 10.449 10.450 10.451 10.452
65
Figure 3.2 Distribution of pairwise spatial distances of sampled ecotypes in meters (m).
We find that soil salinity significantly correlates with elevation (r = 0.38, P < 0.0001, CI =
[0.15, 0.39]), suggesting that elevation is a potential factor that influences soil salinity
variability. We implemented a clustering analysis (fpc package in R) to estimate the number
of clusters along the elevation gradient (Figure 3.3) and identified two distinct clusters (P <
0.001), where high elevation patches (>10 ft.) show a strong positive relationship with soil
salinity levels (r = 0.54, P <0.05), while low elevation patches (<10ft.) show a modest
negative correlation with soil salinity levels (r = -0.25, P < 0.05). Consistent with these
Distribution of Pairwise Spatial Distance
Pairwise distance of sampled plants (m)
Frequency
0 100 200 300 400 500
0 200 400 600 800
66
results, similarity matrices of elevation and salinity levels show a modest but significant
positive relationship (Mantel statistic = 0.29, P < 0.001), indicating that soil patches with
similar soil conductivity share comparable topological profiles. We note that because plants
were generally clustered in patches and the salinity and spatial measurements only reflect
locations of where the ecotypes were sampled; therefore, our data does not completely
capture the spatial and topographical patterns influencing soil salinity variability.
Figure 3.3 3D spatial distribution of accessions collected from the field and corresponding salinity
levels. The x, y, z-axis corresponds to latitude, longitude and elevation, respectively. Blue and red
represent saline patches with Low salinity and Moderate to high salinity, respectively.
3.3.3 Flowering time varies along the soil salinity gradient in the field
Because flowering time is a phenotypically fixed response, this may reflect the potential of
locally specialized responses to soil microsites. We tested whether flowering time varies
along a salinity gradient and found that flowering time is modestly correlated with field soil
salinity levels (Pearson correlation r = -0.17, P < 0.05, N=252). We recognize that some
individuals in this environment may be more adapted to their home microsites than others.
Therefore, we expect that individuals that are most adapted to show a stronger relationship
67
between flowering and salinity level than poorly adapted plants. Because reproductive output
is a qualitative measure of fitness and magnitude of adaptability (Levins, 1968; Orians, 1980),
we examined the tails (95
th
percentile) of the pod production distribution in saline
greenhouse conditions to maximize genetic-environment associations: the left tail represents
the low performance groups and the right, high performance group. We find that among the
high performing individuals, flowering time is strongly correlated with soil salinity (r = -0.57,
P< 0.0001, N=50), where ecotypes originating from seeds collected from high saline
microsites tend to flower earlier than those from low saline microsites (Figure 3.4).
Figure 3.4 Correlation between flowering time and soil salinity. Pearson correlation between
flowering time and electrical conductivity readings of the high performing individuals (N=50,
r=0.578, P < 0.001, Adjusted R
2
= 0.31).
The high performance group typically originated from larger seeds relative to the low
performance group (Wilcoxon Rank Sum test, W(95) = 38908, P < 0.05). Furthermore,
2 4 6 8 10 12
60 65 70 75
EC Reading (mS)
Days to Flowering
68
ecotypes originating from higher elevation microsites tend to flower earlier (r = -0.39, P <
0.001, Figure 3.5); this relationship may be due to the significant correlation between
elevation and salinity, where the influence of elevation on flowering time variation may be
driven by this correlation. No such relationships were found in the low performance group.
Figure 3.5 Correlation between flowering time and elevation. Pearson correlation between flowering
time and elevation among high performing individuals (N=50, r=0.39, P < 0.001, Adjusted R
2
=0.15).
3.3.4 Genome sequencing and population structure analysis
In our previous study, we detected patterns of local adaptation in two saline (TN1 and TN8)
and two non-saline populations (TN7 and TN9) (Friesen et al. 2015). Population TN8 is
adapted to a saline environment exhibiting extreme soil salinity variability with no patterns of
population structure (Friesen et al. 2015), which is an ideal situation for associating genetic
0 10 20 30 40
60 65 70 75 80
Elevation (ft.)
Flowering time (days)
Elevation
(ft.)
Days
to
Flowering
69
and phenotypic variation with microenvironmental variation (Pritchard et al. 2000). The
accessions in this study were collected from a larger area surrounding the TN8 population.
To maximize the power of genotype-phenotype-environment association, we sequenced 96
accessions represented in the tails of the phenotypic distribution as described in the previous
section. We report an average of 6 million 90bp reads were mapped to the M. truncatula 4.0
reference genome, resulting in an average of ~2-5X coverage and identified 227,235 SNPs.
We used STRUCTURE to infer population structure and included the four previously
characterized populations (Friesen et al., 2015). Analysis of 500,000 SNPs with model
likelihood saturating at K=2 suggests that these populations originated from two distinct
ancestral populations. Consistent with expectations, this population of 96 accessions exhibits
no population structure and shares similar haplotype blocks with TN8 (Figure 3.6).
Figure 3.6 STRUCTURE plots comparing the Soliman population with four previously
characterized Tunisian populations. TN7 and TN9 are populations collected from non-saline habitats
and TN1 and TN8 are from saline habitats. The Soliman lines were collected from the same site
where we collected the TN8 population.
3.3.5 Selection along microenvironmental gradients
To understand whether allele frequency is shaped by fine-scale environmental variation, we
tested the correlation of the allele frequency distribution with environmental factors using a
multiple regression approach. Since we found no patterns of population subdivision (Figure
3.6), this data set is ideal for regression analyses (De Mita et al., 2013). Each polymorphic site
was represented as a response variable in the model. Since there is a relationship between
70
elevation and salinity readings for flowering time, we included these factors as the predictor
variables in addition to latitude and longitudinal coordinates. Because allele frequency
estimates are biased for small samples sizes, we filtered out polymorphic sites that were
covered by less than 80% of genotypes with a read depth of at least four reads. We evaluated
4,721 SNPs and found 663, 21, 7, and 2 SNPs that are significantly associated with latitude,
longitude, elevation and soil conductivity readings, respectively (FDR<0.05, Figure 3.7).
Figure 3.7 Manhattan plots of the SNPs that are significantly associated with latitude, longitude,
elevation and soil conductivity in the field. Red line indicates a significance threshold of FDR < 0.05.
Allele frequency correlates strongly with latitude, with nearly 50% of these polymorphisms
causing amino acid substitutions, while only ~20% of alleles associated with longitude cause
71
functional changes. Six out of the seven elevation-associated SNPs fall within gene coding
regions and four are non-synonymous SNPs (Figure 3.7). Interestingly, many of these genes
are associated with oxidative and salt stress response, disease resistance and nodulation in
legumes (Table 3.3). Salinity on the other hand, had the least number of significantly
associated polymorphisms with none causing amino acid changes.
Table 3.3 Annotations of genes associated with significant SNPs that correlate with elevation in the
field along with corresponding SNP effects. (S: synonymous, NS: nonsynonymous)
In our previous study, we examined two saline and two non-saline populations that
consisted of ten genotypes per population and identified functional polymorphisms
associated with earlier flowering in saline environments (Friesen et al., 2015). However, with
small sample sizes and reduced power, we were only able to test specific regions exhibiting
the most differentiation between saline and non-saline soil types. In this study, we sampled
96 accessions, which increased our power to capture whole genome genetic differences
associated with flowering. To understand whether there is a genetic basis for fine-scale
flowering time differences along the salinity gradient in the field, we estimated F
st
to identify
genetic differences along this gradient. We tested for regions of differentiation based on
72
flowering time differences associated with home soil microsites in the high performance
group (N = 50). Subpopulations were partitioned according to low (0-4 mS) versus high
saline (4-12 mS) microsites (Shirokova, 2000; Landon, 2014). Since linkage disequilibrium
among Tunisian M. truncatula is ~10kb (Friesen et al., 2015), we estimated mean F
st
of non-
overlapping 10kb windows for 227,236 SNPs. Because large F
st
values were generally
associated with low SNP density, we ran permutation to identify a SNP density threshold for
each window. We sampled individuals from the pool of 96 accessions without replacement
and generated 1000 simulated F
st
estimates to determine the SNP density threshold within
10kb windows based on a 95% confidence interval. The average number of SNPs within a
10kb window is 23 (std. dev = 1.39) with a threshold cutoff of at least 6 SNPs per 10kb
window. The mean F
st
across all windows is 0.0052 and the estimates per 10kb window
ranged from 0-0.49.
We permuted the data to assess the significance of F
st
estimates greater than 0.1 (95
th
percentile of average F
st
values, Figure 8). Similar to the permutations done for the SNP
density threshold described above, we compared the permuted F
st
estimates to the observed
F
st
estimates to assign significance to each 10kb window (FDR < 0.05). We identified 561
windows that fell above the 95
th
percentile of average F
st
values (Figure 3.8).
73
Figure 3.8 Distribution of average F st values. Values correspond to non-overlapping 10kb windows
between low saline and high saline patches for high performance individuals that show correlated
flowering time (N=50). The red line marks the 95
th
percentile of values (average F st = 0.115).
3.4 Discussion
In this study we evaluate the patterns of micro-scale adaptation in a single selfing population
inhabiting a patchy saline environment. We test whether this environment favors
phenotypically fixed or plastic responses in flowering and performance related traits. This
population exhibits no patterns of genotype-environment interactions in any of the traits
tested and displays patterns indicative of locally specialized ecotypes; these patterns are most
apparent in flowering time. Flowering time has been previously characterized as an
Fst of 10kb windows
Frequency
-0.1 0.0 0.1 0.2 0.3 0.4 0.5
0 1000 2000 3000 4000 5000
74
important adaptive response in saline environments (Friesen et al., 2015) and our results are
consistent with these findings. We find that flowering is a phenotypically fixed trait
associated with high performance and varies predictably along a salinity gradient. Genome
sequencing indicates differential selection among soil microsites that appears to facilitate
genetic differentiation and consequently the maintenance of genetic and phenotypic
variation.
Gillespie & Turelli (1989) hypothesized that genotype-environment interaction is a major
source of genetic variability in spatially and temporally varying environments, where
differential selection for alternative alleles among microenvironments facilitates the
maintenance of genetic variation (Gillespie and Turelli, 1989). However, we did not detect
significant genotype-environment interactions in flowering and performance related traits
(i.e. leaf size, leaf number, number of branches and pod number), suggesting that phenotypic
plasticity is unable to evolve in this population. One explanation for the absence of genetic
variation in phenotypic plasticity is that selection for an optimal or advantageous level of
plasticity may be too slow and ineffective in populations exhibiting low migration and low
genetic variability (Via and Lande, 1985).
Furthermore, Gillespie & Turelli (1989) concluded that genotype-environment interaction is
a strong mechanism maintaining genetic variation by demonstrating that the mean fitness of
a genotype is a function of increasing number of heterozygote sites (Gillespie and Turelli,
1989). They argued that phenotypic plasticity or so called developmental homeostasis and
environmental buffering is a result of increased levels of heterozygosity. In light of this point
of view, the maintenance of genetic variation by way of heterozygote selection may be more
75
relevant in outcrossing rather than selfing populations. Because inbreeding in selfers is
expected to significantly reduce heterozygosity and genetic diversity (Wright, 1921; Mitton,
1993), the evolution of adaptive plasticity may be restricted in highly selfing populations.
Although our data suggests that adaptive phenotypic plasticity is unable to evolve in this
predominantly selfing population, we do not have sufficient data to validate this observation
and provide an explanation for the observed patterns. Therefore, further studies examining
mating systems and its influence on patterns of microenvironmental adaptation and
maintenance of genetic variation is necessary to address this evolutionary question.
Annual plants depend on environmental cues to initiate the timing of flowering to ensure
reproductive success (i.e. temperature, day length and soil moisture) (Schmitt, 1983;
RATHCKE and Lacey, 1985; Galloway and Burgess, 2012). Annuals are sensitive to seasonal
variability with strong selection for appropriate flowering time response and this has been
demonstrated in M. truncatula as well as rice, Arabidopsis, and Mimulus (Hall and Willis, 2006;
Franks et al., 2007; Izawa, 2007; Galloway and Burgess, 2012). M. truncatula is a short lived
annual: seeds germinate in the winter, plants flower in the spring and begin to senesce in the
summer. Flowering time has been linked to reproductive success and phenotypic
differentiation in natural M. truncatula populations with strong selection for earlier flowering
in saline environments (Friesen et al., 2015). In this study, we find that saline origin ecotypes
demonstrating high performance and reproductive output tend to flower earlier (Table 3.2).
Soil salinity levels begin to increase in the spring and peak in the summer; plants adapted to
saline environments tend to flower early, a potential salt avoidance mechanism to escape
toxic levels of salt to ensure survival to reproduction (Friesen et al., 2015). Similar flowering
76
time patterns were seen in a separate experiment assessing parental environmental effects
using the same populations (Moriuchi et al., 2014).
Micro-scale adaptation to patch conditions will depend on seed dispersal rates and the
predictability of environmental cues. Dispersal rates should be high enough to allow
distribution of genotypes across the landscape, but low enough to maintain cross-
generational environmental predictability. Under migration-selection balance, even under
relatively high dispersal rates between patches, genetic constraints can lead to adaptation to
patch conditions and phenotypic divergence (Scheiner, 1998). M. truncatula pods are spiny
and pod dispersal is likely mediated by grazing animals that may facilitate pod dispersal along
the landscape, but a previous study of M. truncatula ecotypes detected low within population
migration between patches (Bonnin et al., 2001). Because self-fertilization permits higher
transmission rates of adaptive alleles and provides a barrier against gene flow (Antonovics,
1968; GOLENBERG, 1987), selfing populations are likely more successful in colonizing
environments with microscale variation and exhibit strong patterns of gene by environment
correlation. In this study we find that flowering time is strongly correlated with salinity levels
among the individuals that are most adapted within this population; this could potentially be
an outcome of low migration between patches resulting in high parent-offspring
environment correlations.
One caveat of this study is that we cannot disentangle parental environmental effects on
plant responses as seeds collected from the field were grown directly in the greenhouse
without a generation of common growth environments. We did observe a large effect of
seed size on several traits, which suggests one possible mode of transmission of parental
77
environmental effects. However, in an examination of parental environmental effects in
four populations from Northern Tunisia, including the TN8 population sampled here, we
found little effect of seed size on traits when we specifically manipulated soil salinity in the
parental environment (Castro et al., 2013; Moriuchi et al., 2014). Consistent with these
previous results, seed size did not correlate with home microsite where parental plants were
found, suggesting minimal parental environmental effect on seed traits. Furthermore,
because flowering time is not responsive to greenhouse salt treatments, and is not affected
by parental saline exposure (Moriuchi et al., 2014), we are confident that the strong
correlation observed between flowering time and soil conductivity in the field is likely due to
adaptation to soil patches in the field.
In summary, this study has identified that potential of maintenance of genetic variation
through microgeographical adaptation, where selection differential selection among soil
patches can facilitate divergent selection on flowering time responses. We find a lack of
genotype-environment interactions in flowering time and performance related traits,
suggesting that phenotypic plasticity is unable to evolve within this population. This may be
explained by genetic constraints seen in selfing populations, where inbreeding causes a
reduction in heterozygosity and consequently genetic variation. However, flowering time is
phenotypically fixed and correlates with the salinity gradient in the field, indicating variability
in salinity levels may be driving differential selection leading to adaptation to local patch
conditions. Indeed, we find significant population differentiation associated with flowering
time and soil salinity differences.
78
3.5 Materials and Methods
3.5.1 Field collection and greenhouse experiment
On June 5 and 6 2009, we sampled 252 natural accessions from a natural saline site in
Soliman, located along the northern coast of Tunisia (Figure 3.1). Characteristic of salt
affected land, this field site exhibits extreme variability in soil salinity levels that vary both
spatially and temporally. Pods were collected from individual plants and GPS coordinates
and soil electrical conductivity readings were recorded at the base of each plant. Electrical
conductivity readings were taken as a measure of microsite soil salinity levels experienced by
the plant. To examine phenotypic responses to salinity, six seeds per line (3 seeds per
treatment) were individually weighed to the nearest 0.0001 mg and scarified. Seeds were
sown on 2/15/10 in sterile horticulture sand in 656 ml DeePots in a UC Davis greenhouse
and treated with 0.1X Fahräeus nutrient solution supplemented with 3mM KNO
3
spiked
with either 0mM NaCl or 100mM NaCl twice weekly. In addition to treatments, plants were
watered twice a day to prevent drying and randomized once a month. First census for
growth and developmental traits were taken one month (3/13/10) from the start of the
experiments and a second census was conducted on 3/31/10. We recorded data on first
flowering and when plants died. First pods were collected in mid-late July (7/16/10 –
7/29/10). Plants that were dead by early august were harvested (~8/13/10) with all other
plants harvested in late August early September, which included all mature pods.
3.5.2 Genome sequencing and library construction
To examine the genetic basis of adaptation to micro-environmental variation in a single
population, we sequenced a subset of the 252 lines. To maximize genotype-phenotype-
environmental associations, we used a stratification method to select individuals for
79
sequencing. 96 individuals were selected from the tails of two distributions based on pod
production: pod production in saline greenhouse conditions and pod production in saline
condition relative to non-saline conditions. This resulted in a bulk segregant or
subpopulations of individuals that show high performance and low performance in saline
conditions. These individuals should show the greatest genetic difference in traits associated
with performance in contrasting salinity microsites.
A total of 96 accessions were selected for Illumina sequencing. DNA was isolated from leaf
tissues and fragmented using dsDNA Shearase (Zymo Research Product #E2018-200). The
fragments were blunt-end repaired using a Quick Blunting Kit (NEB #E0542S) and a single
A to the blunt end using Klenow Fragment 3’-5’ exo-nuclease (NEB Product #M0212L,
Ipswich, MA). Ilumina adaptors were ligated to the DNA fragments using the Epicentre
Fast-Link DNA Ligation Kit (Product #LK6201H, Madison, WI). Fragments between 300-
500bp were size selected by agarose gel and samples were indexed and enriched. The
libraries were tagged using 96 barcoded indices and paired-end 90bp reads were obtained
from each accession resulting in 2-5X coverage. Reads were aligned to the M. truncatula 3.5.1
reference genome and genotyping was done using GATK Unified Genotyper with a custom
quality threshold based on Sanger sequences from four loci to obtain a false discovery rate of
<5%.
3.5.3 Population structure analysis
We used STRUCTURE (Falush et al., 2003; Hubisz et al., 2009) to infer population structure
based on 500,000 loci. STRUCTURE was run under the admixture and correlated allele
frequency model. Ten independent runs of 10,000 burn-in MCMC iterations following by
80
50,000 iterations were performed for 2 to 12 clusters (k=2 to 12). Results were inspected
with STRUCTURE HARVESTER (Evanno et al., 2005).
3.5.4 Statistical analysis on plant traits
To understand plant responses and performance in saline conditions in the greenhouse, we
tested the influence of line, treatment, and seed mass on plant traits using ANOVA
(implemented by the ‘car’ package in R). We considered leaf size, leaf number, branch
number, flowering time and pod production in our analysis. Model fitting was assessed with
Akiake’s information criterion. For each model we considered interactions between main
effects, and included seed weight as a covariate. Natural log and square root transformation
were implemented to satisfy the model assumptions of normality and homoscedasticity.
Phenotypic plasticity is indicated by a significant treatment effect and genetic differences
between accessions and genetic variation in plasticity are indicated by significant line by
treatment interaction, respectively.
3.5.5 Fst estimates of 10kb windows
In order to associate genetic differences associated with flowering time divergence relative to
salinity microsite variation in the field, we estimated average F
st
over 10kb sliding windows
using Weir and Cockerham’s weighted estimates (Weir & Cocherham, 1984) implemented by
Vcftools.
3.5.6 Multiple regression on allele frequency
To evaluate whether selection can operate on microenvironmental variation, we used a
logistic regression model to test the influence of environmental factors on allele frequencies.
81
Logistic regression models are powerful approaches for identifying loci involved in
adaptation to heterogeneous environments (Joost et al., 2007; De Mita et al., 2013), and have
been shown to perform particular well in selfing populations that exhibits no populations
structure and where sampling location are represented by a single individual(De Mita et al.,
2013). We implemented ordered logistic regression model in R using the MASS package.
Allele frequencies were calculated to determine minor and major alleles: allele frequencies <
0.5 were considered minor alleles. At each polymorphic position, the genotypes were coded
as categorical variables. SNP categories were considered response variables with latitude,
longitude, elevation and salinity levels as predictor variables. Since ordered logistic regression
assumes that the distance between each category is equivalent (i.e. proportional odds
assumption), we implemented Harrell (2001) graphical method for assessing the proportional
odds assumption.
3.6 References
Al-Hiyaly SA , McNeilly T, Bradshaw AD (1988). The effects of zinc contamination from
electricity pylons - evolution in a replicated situation. New Phytologist 110: 571–580.
Antonovics J (1968). Evolution in Closely Adjacent Plant Populations .v. Evolution of Self-
Fertility. Heredity (Edinb) 23: 219–&.
Antonovics J, Bradshaw AD (1968). Evolution in closely adjacent plant populations. Heredity
(Edinb).
Bonnin I, Ronfort J, Wozniak F, Olivieri I (2001). Spatial effects and rare outcrossing events
in Medicago truncatula (Fabaceae) - Bonnin - 2001 - Molecular Ecology - Wiley Online
Library. Molecular Ecology.
Bourne EC, Bocedi G, Travis JMJ, Pakeman RJ, Brooker RW, Schiffers K (2014). Between
migration load and evolutionary rescue: dispersal, adaptation and the response of
spatially structured populations to environmental change. Proceedings of the Royal Society B:
Biological Sciences 281.
Bradshaw AD, McNeilly TS, Gregory R (1965). Bradshaw: Industrialization, evolution and
82
the developmen... - Google Scholar. Ecology and the industrial society.
Castro BM, Moriuchi KS, Friesen ML, Badri M, Nuzhdin SV, Strauss SY, et al. (2013).
Parental environments and interactions with conspecifics alter salinity tolerance of
offspring in the annual Medicago truncatula (P Klinkhamer, Ed.). J Ecol 101: 1281–1287.
Charlesworth D, Morgan MT, Charlesworth B (1992). The effect of linkage and population
size on inbreeding depression due to mutational load. Genet Res 59: 49–61.
Chaulet E, Prosperi JM (1994). Genetic diversity of a collection of Medicago trunculata
Gaernt from Algeria. EUCARPIA: 15–18.
Conover DO, Duffy TA, Hice LA (2009). The covariance between genetic and
environmental influences across ecological gradients: reassessing the evolutionary
significance of countergradient and cogradient variation. Ann N Y Acad Sci 1168: 100–
129.
De Mita S, Thuillet A-C, Gay L, Ahmadi N, Manel S, Ronfort J, et al. (2013). Detecting
selection along environmental gradients: analysis of eight methods and their
effectiveness for outbreeding and selfing populations. Molecular Ecology 22: 1383–1399.
Evanno G, Regnaut S, Goudet J (2005). Detecting the number of clusters of individuals
using the software STRUCTURE: a simulation study. Molecular Ecology 14: 2611–2620.
Falconer DS, Mackay TFC (1996). Introduction to Quantitative Genetics, 4 edn. Trends in
Genetics.
Falush D, Wirth T, Linz B, Pritchard JK, Stephens M (2003). Traces of Human Migrations
in Helicobacter pylori Populations. Science.
Franks SJ, Sim S, Weis AE (2007). Rapid evolution of flowering time by an annual plant in
response to a climate fluctuation. Proc Natl Acad Sci USA 104: 1278–1282.
Friesen ML, Wettberg EJB, Badri M, Moriuchi KS, Barhoumi F, Chang PL, et al. (2015). The
Ecological and Genomic Basis of Salinity Adaptation in Tunisian Medicago truncatula.
BMC Genomics in press: 1–47.
Galloway LF, Burgess KS (2012). Artificial selection on flowering time: influence on
reproductive phenology across natural light environments. J Ecol 100: 852–861.
Gillespie JH, Turelli M (1989). Genotype-environment interactions and the maintenance of
polygenic variation. Genetics 121: 129–138.
GOLENBERG EM (1987). Estimation of Gene Flow and Genetic Neighborhood Size by
Indirect Methods in a Selfing Annual, Triticum-Dicoccoides. Evolution 41: 1326–1334.
Griffith TM, Watson MA (2005). Stress avoidance in a common annual: reproductive timing
is important for local adaptation and geographic distribution. Journal of Evolutionary Biology
18: 1601–1612.
83
HAJRASULIHA S, BANIABBASSI N, METTHEY J, NIELSEN DR (1980). Spatial
Variability of Soil Sampling for Salinity Studies in Southwest Iran. Irrigation Science 1: 197–
208.
Hall MC, Willis JH (2006). Divergent selection on flowering time contributes to local
adaptation in Mimulus guttatus populations. Evolution 60: 2466–2477.
Holsinger KE (1986). Dispersal and Plant Mating Systems - the Evolution of Self-
Fertilization in Subdivided Populations. Evolution 40: 405–413.
Hubisz MJ, Falush D, Stephens M, Pritchard JK (2009). Inferring weak population structure
with the assistance of sample group information. Molecular Ecology Resources 9: 1322–1332.
Izawa T (2007). Adaptation of flowering-time by natural and artificial selection in
Arabidopsis and rice. Journal of Experimental Botany 58: 3091–3097.
Joost S, Bonin A, Bruford MW, Després L, Conord C, Erhardt G, et al. (2007). A spatial
analysis method (SAM) to detect candidate loci for selection: towards a landscape
genomics approach to adaptation. Molecular Ecology 16: 3955–3969.
Kawecki TJ, Ebert D (2004). Conceptual issues in local adaptation. Ecol Letters 7: 1225–1241.
Landon JR (2014). Booker tropical soil manual: a handbook for soil survey and agricultural
land evaluation in the tropics and subtropics.
LEVENE H (1953). Genetic Equilibrium When More Than One Ecological Niche is
Available on JSTOR. American Naturalist 87: 331–333.
Levins R (1968). Evolution in Changing Environments. Princeton University Press: Princeton,
New Jersey.
Loveless MD, Hamrick JL (1984). JSTOR: Annual Review of Ecology and Systematics, Vol.
15 (1984), pp. 65-95. Annual Review of Ecology and Systematics.
Mills KE, Bever JD (1998). Maintenance of Diversity within Plant Communities: Soil
Pathogens as Agents of Negative Feedback. Ecology 79: 1595–1601.
Mitton JB (1993). The natural history of inbreeding and outbreeding. Oxford University Press.
Miyamoto S, Chacon A, Hossain M, Martinez I (2005). Soil salinity of urban turf areas
irrigated with saline water. Landscape and Urban Planning 71: 233–241.
Moriuchi KS, Friesen ML, Cordeiro MA, Badri M, Vu WT, Main B, et al. (2014). The role of
parental and offspring environment on population responses of Medicago truncatula to soil salinity. in
review: American Journal of Botany.
Nei M (1972). Genetic Distance between Populations on JSTOR. American Naturalist.
Nordborg M (1997). Structured coalescent processes on different time scales. Genetics 146:
84
1501–1514.
Ollerton J, Lack A (1998). Relationships between flowering phenology, plant size and
reproductive success in Lotus corniculatus (Fabaceae). Plant Ecology 139: 35–47.
Olsson K, Ågren J (2002). Latitudinal population differentiation in phenology, life history
and flower morphology in the perennial herb Lythrum salicaria. Journal of Evolutionary
Biology 15: 983–996.
ORIANS GH (1980). Orians: Habitat selection: general theory and application... - Google
Scholar. Elsevier: 49–66.
Palumbi SR (1994). Genetic Divergence, Reproductive Isolation, and Marine Speciation.
Annu Rev Ecol and Syst 25: 547–572.
RATHCKE B, Lacey EP (1985). Phenological Patterns of Terrestrial Plants. Annual Review of
Ecology and Systematics 16: 179–214.
Rousset F (1997). Genetic differentiation and estimation of gene flow from F-statistics under
isolation by distance. Genetics 145: 1219–1228.
Scheiner SM (1998). The genetics of phenotypic plasticity. VII. Evolution in a spatially-
structured environment. Journal of Evolutionary Biology 11: 303–320.
Schmalhausen II (1949). Factors of evolution.
Schmitt J (1983). Individual Flowering Phenology, Plant Size, and Reproductive Success in
Linanthus-Androsaceus, a California Annual. Oecologia 59: 135–140.
Schmitt J, Gamble SE (1990). The Effect of Distance from the Parental Site on Offspring
Performance and Inbreeding Depression in Impatiens capensis: A Test of the Local
Adaptation Hypothesis. Evolution 44: 2022.
Shi Z, Li Y, Wang R, Makeschine F (2005). Assessment of temporal and spatial variability of
soil salinity in a coastal saline field. Environmental Geology 48: 171–178.
Shirokova Y (2000). Use of Electrical Conductivity Instead of Soluble Salts for Soil Salinity
Monitoring in Central Asia - Springer. Irrigation and Drainage Systems 14: 199–206.
Siol M, Prosperi JM, Bonnin I, Ronfort J (2008). How multilocus genotypic pattern helps to
understand the history of selfing populations: a case study in Medicago truncatula.
Heredity (Edinb) 100: 517–525.
Slatkin M (1985). Gene flow in natural populations. Annual Review of Ecology and Systematics:
393–430.
Snaydon RW (1970). Rapid Population Differentiation in a Mosaic Environment. I. The
Response of Anthoxanthum odoratum Populations to Soils. Evolution 24: 1–14.
85
Stinchcombe JR, Dorn LA, Schmitt J (2004). Flowering time plasticity in Arabidopsis
thaliana: a reanalysis of Westerman & Lawrence (1970). Journal of Evolutionary Biology 17:
197–207.
Sultan SE, Spencer HG (2002). Metapopulation Structure Favors Plasticity over Local
Adaptation. The American Naturalist 160: 271–283.
TURESSON G (1922). j.1601-5223.1922.tb02727.x. Hereditas 3: 1–14.
Via S, Lande R (1985). Genotype-environment interaction and the evolution of phenotypic
plasticity. Evolution: 505–522.
WADLEIGH CH, FIREMAN M (1948). Salt Distribution Under Furrow and Basin
Irrigated Cotton and Its Effect on Water Removal. Soil Science Society of America Proceedings
13: 527–530.
Walter C, McBratney AB, Douaoui A, Minasny B (2001). CSIRO PUBLISHING - Soil
Research. Aust J Soil Res 39: 259.
Wright S (1921). Systems of Mating. I. the Biometric Relations between Parent and
Offspring. Genetics 6: 111–123.
Wright S (1943). Isolation by distance. Genetics 28: 114.
Wright SI, Kalisz S, Slotte T (2013). Evolutionary consequences of self-fertilization in plants.
Proceedings of the Royal Society B: Biological Sciences 280: 20130133.
86
Chapter 4
Genetic variation in transgenerational plasticity of
germination and seed transcriptome
The following chapter has been accepted for publication in BMC Evolutionary Biology:
Wendy T. Vu, Peter L. Chang, Ken S. Moriuchi, Maren L. Friesen
Genetic variation in transgenerational plasticity of the seed transcriptome and
offspring germination response to salinity stress in Medicago truncatula.
BMC Evolutionary Biology 2015, in press.
4.1 Abstract
Transgenerational plasticity, a form of parental environmental effects, provides phenotypic
variation that plays an important role in adaptation. For plants, the timing of seed
germination is critical for offspring survival in stressful environments, as germination timing
can alter the environmental conditions a seedling experiences. Stored seed transcripts are
important determinants of seed germination, but have not previously been linked with
transgenerational plasticity of germination behavior. In order for transgenerational plasticity
to play an ongoing role in adaptation, there must be genetic variation for this trait
segregating within species. In this study we used RNAseq and growth chamber experiments
of the model legume M. trucantula to test whether parental exposure to salinity stress
influences the expression of stored seed transcripts and early offspring traits.
87
We detected genotype-dependent parental environmental effects (transgenerational
plasticity) on the expression levels of stored seed transcripts, seed size, and germination
behavior of four M. truncatula genotypes. Across all genotypes, more than 50% of the
transcripts detected in the mature, ungerminated seed transcriptome were annotated as
regulating seed germination. Gentoypes TN7.22 and TN8.22, showed increased seed size in
response to parental exposure to salinity stress, but no parental environmental influence on
germination timing. In contrast, genotypes TN1.13 and TN1.15 showed no seed size
differences across contrasting parental conditions but displayed transgenerational plasticity
for germimation timing, with significantly delayed germination in saline conditions when
parental plants were exposed to salinity. We conducted a coexpression network analysis and
found significant sub-networks derived from salt responsive transcripts in TN1.13 and
TN1.15 that are involved in post-transcriptional regulation of the germination pathway.
Consistent with the delayed germination response to saline conditions in these genotypes, we
found significant clustering of correlated genes associated with dormancy and up-regulation
of abscisic acid (ABA).
Our results demonstrate genetic variation in transgenerational plasticity within M. truncatula
and show that parental exposure to salinity stress influences the expression of stored seed
transcripts, seed weight, and germination behavior. Observed transgenerational plastic
responses to salinity may be adaptive as soil salinity levels decrease during the germination
season, but tests of adaptation requires future studies. Furthermore, we observed that
parental environmental effects were due to differential seed provisioning or seed transcript
expression, suggesting that different mechanisms may play a role in germination and early
seedling characters in M. truncatula. We identified potential genes/transcripts that will be the
88
focus of future studies on salinity adaptation in M. truncatula. Our coexpression analysis
suggests crosstalk between stress and dormancy pathways that influence germination timing
in some genotypes within natural populations. Furthermore, we show that the parental
environment influences gene expression to modulate biological pathways that are likely
responsible for offspring germination responses to salinity stress.
4.2 Introduction
Transgenerational plasticity occurs when the parental environment influences offspring
development and responses to environmental conditions in the absence of genetic changes.
While transgenerational plasticity is not necessarily adaptive, it is predicted to influence the
rate of adaptation by changing the strength and direction of responses to selection in the
offspring generation (Kirkpatrick and Lande, 1989; Hoyle and Ezard, 2012; Friesen et al.,
2015). Evolutionary theory shows that transgenerational plasticity can sometimes be an
adaptive mechanism that increases long-term fitness under environmental heterogeneity
(Marshall, 2008; Donohue, 2009b; Dyer et al., 2010; Moriuchi et al., 2014) and extreme
environmental shifts (Mousseau et al., 2009; Roberts et al., 2012; Hoyle and Ezard, 2012).
Parental exposure to predation in three-spine sticklebacks and crickets have been shown to
adaptively influence offspring anti-predator behavior (Kirkpatrick and Lande, 1989; Storm
and Lima, 2010; Kozak and Boughman, 2012; Roberts et al., 2012; Hoyle and Ezard, 2012)
and defense phenotypes of wild radish progeny have been correlated with parental exposure
to herbivory (Agrawal, 2002; Marshall, 2008; Donohue, 2009b; Dyer et al., 2010; Robertson et
al., 2010). However, in situations where environments are highly unpredictable (Van Dam
and Baldwin, 2001; Shannon et al., 2003; Reed et al., 2010) or conditions fall outside the
89
adaptive range of a population (Shannon et al., 2003; Galloway and Etterson, 2007; Sultan et
al., 2009), transgenerational plasticity can also be maladaptive. In order for transgenerational
plasticity to itself evolve, it must be heritable and exhibit variation between genotypes. There
is evidence that transgenerational plasticity is genetically based in field studies examining the
influence of the parental environment on offspring response and performance relative to
contrasting parental environments (Lacey and Herr, 2000; Galloway and Etterson, 2007;
Bassel et al., 2008; Mousseau et al., 2009; Hoyle and Ezard, 2012). Several studies in plants
have reported genotypic differences in adaptive transgenerational plasticity (Galloway, 2001;
Bader and Hogue, 2003; Galloway and Etterson, 2007; Castro et al., 2013), where genotypes
differ in the type of transgenerational mechanisms responsible for the transmission of
adaptive environmental cues to their offspring.
Plants have two well-characterized processes that mediate transgenerational plasticity in
offspring performance and response: resource provisioning as reflected in seed size (Roach
and Wulff, 1987; Rossiter, 1996)and seed dormancy/germination pathways (Biere, 1991;
Donohue, 2009a). Seed size is often linked directly to performance variation through growth
rate and competitive ability (Eriksson, 1999; Chacón and Bustamante, 2001), while the
dormancy pathways determine germination timing—an important life history trait that
determines the environment experienced by the developing plant (Donohue et al., 2010). In
addition to dormancy pathways, seed size is often a determinant of germination behavior in
some but not all species (Crawley and Nachapong, 1985; Schmid and Dolt, 1994; Moshatati
and Gharineh, 2012). In Pinus pinaster, parental environmental effects on seed mass partially
explained variation in germination timing between contrasting parental environments
(Cendán et al., 2011).
90
Dormancy is an adaptive life history trait to seasonally unfavorable environmental conditions
(Williams and Harper, 1965; Donohue et al., 2010). Dormancy is established during seed
development on the parental plant (Kendall et al., 2011); therefore, this trait is likely
programmed during seed development. There are a variety of mechanisms that have evolved
to mediate seed dormancy in angiosperms. Physical seed dormancy is mediated by the seed
coat (“hardseededness” or “seed hardiness”) (Baskin and Baskin, 1998; Russi et al., 2008;
Bolingue et al., 2010). The seed coat is a maternal tissue that is made of a waxy, hydrophobic
tegument that prevents the uptake of water and oxygen required for germination (Garcia et
al., 2006; Taiz and Zeiger, 2010) and is likely under the control of the maternal genotype
(Roach and Wulff, 1987; Lacey et al., 1997). Another mechanism is physiological dormancy,
which is represented by two classes: primary and secondary dormancy. Primary dormancy is
maintained by the accumulation of phytohormone abscisic acid (ABA) during seed
maturation to prevent precocious seed germination (Bolingue et al., 2010), and requires a
period of after-ripening before seeds have the capacity to germinate under favorable
conditions. Secondary dormancy, on the other hand, is the re-induction of dormancy by
after-ripened non-dormant seeds in response to certain environmental conditions (Baskin
and Baskin, 1978; Bewley, 1997; Baskin and Baskin, 1998; Penfield and Springthorpe, 2011),
particularly unfavorable ones (Roach and Wulff, 1987; Lacey et al., 1997). Imbibed after-
ripened seeds of Arabidopsis thaliana can induce secondary dormancy under certain
temperature regimes through changes in the expression of dormancy related transcripts
(Cadman et al., 2006; Finch-Savage et al., 2007), suggesting that transcriptional regulation can
mediate delayed germination response to offspring environmental condition.
91
Transcripts stored in mature seeds play a critical role in regulating seed germination and
dormancy in a variety of plant species (Almoguera and Jordano, 1992; Hoecker et al., 1995;
Jones et al., 1997; Bassel et al., 2008; Penfield et al., 2010; Bassel et al., 2011). Work in A.
thaliana using chemical inhibitors of transcription and translation demonstrated that stored
seed transcripts are both necessary and sufficient for seed germination (Rajjou, 2004);
variation in stored seed transcript expression of VIVIPAROUS was related to the length of
time required for seeds to break dormancy in Avena fetua (Jones et al., 1997). Stored seed
transcripts, like other compounds important for early seedling establishment, are deposited
after embryogenesis and during seed maturation (Harada, 1997; Ruuska et al., 2002) and
previous studies have found that expression profiles of stored seed transcripts respond to
the parental environment (Almoguera and Jordano, 1992; Kendall et al., 2011).
Medicago truncatula, a member of the Fabaceae family, is primarily a selfing annual legume
native to the Mediterranean region and found naturally occurring in both saline and non-
saline habitats (Badri et al., 2007; Lazrek et al., 2009; Friesen et al., 2010; 2015). In saline
habitats, salt accumulates at the surface soil during the summer, with soil salt concentrations
peaking during the first rain from the summer drought and then dropping as additional rains
leach salt from the soil surface (Smith and Stoneman, 1970; Nichols et al., 2008). One
mechanism by which pasture legumes adapt to saline environments is by delaying
germination to avoid high salt concentrations early in the rainy season (Nichols et al., 2009).
Along with other Medicago species, M. truncatula is characterized to exhibit both seed coat-
imposed physical dormancy and physiological primary dormancy (Crawford et al., 1989;
Baskin and Baskin, 1998; Patanè and Gresta, 2006; Gallardo et al., 2006; Garcia et al., 2006),
with ABA playing an important role in the latter (Gallardo et al., 2006). Although M.
92
truncatula exhibits non-deep primary dormancy (several weeks/months of after-ripening is
sufficient to remove dormancy), it is unclear whether secondary dormancy occurs in M.
truncatula seeds (Gallardo et al., 2006; Bolingue et al., 2010). While the M. truncatula
transcriptome of developing and dry, mature seeds has been previously characterized
(Benedito et al., 2008), it is currently unknown whether the parental environment influences
stored seed transcripts and whether these transcripts play a role in seed dormancy or delayed
germination responses.
In this study, we explore the molecular mechanisms that are influenced by the parental
environment to facilitate offspring transgenerational responses. We use wild genotypes of
the model legume M. truncatula from naturally saline and non-saline habitats to explore
parental environmental effects on the expression of stored seed transcriptome, which could
potentially mediate transgenerational plasticity in response to salinity stress. We focus on
seed dormancy and identify genes and pathways potentially responsible for transgenerational
plasticity of germination behavior. We test the hypothesis that M. truncatula shows genetic
variation for transgenerational plasticity of germination timing under parental salinity stress.
Using next-generation sequencing, we assess the influence of parental salt exposure on the
stored seed transcriptome and determine whether parental environmental effects on seed
transcript expression differ among genotypes. Finally, we identify novel candidate biological
pathways influenced by the parental environment to facilitate offspring transgenerational
plastic germination behavior.
93
4.3 Materials and Methods
Self-fertilized seeds derived from four inbred Tunisian genotypes of M. truncatula, TN1.13,
TN1.15, TN7.22, and TN8.22, were used to measure transgenerational plasticity (parental
environmental effects) in early developmental phenotypes. These genotypes are a subset of a
larger collection of 39 genotypes derived from two saline (i.e., Enfidha:TN1 and
Soliman:TN8) and two non-saline (i.e., El Kef:TN7 and Bulla Regia:TN9) populations (Badri
et al., 2007; Lazrek et al., 2009; Friesen et al., 2010). TN1.13 and TN1.15 originate from the same
saline source population (TN1), TN8.22 is from the TN8 saline population (TN8), and
TN7.22 is from the TN7 non-saline population (Friesen et al., 2015). These genotypes were
chosen from a population scale parental environmental effects experiment of all 39 Tunisian
genotypes in UC Davis (Moriuchi et al., 2014). TN1.13 and TN1.15 were chosen because they
displayed significant transgenerational plasticity, while TN8.22 and TN7.22 were selected
because they exhibited no patterns of parental environmental influence on offspring
germination behavior.
4.3.1 Parental and offspring environment
During the parental generation, seeds from each genotype were grown in a 2:1 sterile
horticultural sand: UC Davis soil mix in 4cm diameter/20cm long cone-tainers (Steuwe)
under ambient conditions at UC Davis. Two weeks after germination, salt treatments were
initiated by treating half of the plants with a Fahraeus solution with either 0 mM NaCl or
100 mM NaCl (i.e., parental environment). 100 mM NaCl is within the range of salinity
observed in the field, which is sufficient to cause stress but not extreme mortality. For the
offspring generation, ten seeds from each genotype and parental environment were planted
in growth chambers at USC and the conditions were set at 16/8 hour day/night cycle with
94
temperatures at 13°C and 18°C (similar to the growth conditions experienced by the parental
plants). Seeds were after-ripened for one year in dry conditions before they were used for the
offspring germination experiment. Single, unvernalized seeds were weighed and sandpaper
scarified and planted into the same growth medium as the parental plants. Pots were fully
randomized every two weeks until the end of the experiment and seeds were immediately
treated with Fahraeus nutrient solution spiked with either 0 mM or 100 mM NaCl and
subsequent treatments were done twice a week. We recorded the timing of germination,
unifoliate development, and first trifoliate development. Plants were surveyed daily and seeds
were considered germinated when cotyledons were fully expanded. All plants were grown to
senescence and pods were collected as they naturally matured on each plant.
4.3.2 Transcriptome library construction
To test the effects of the parental environment on seed transcripts, we constructed Illumina
Solexa sequencing libraries of 24 self-fertilized dry, ungerminated seeds derived from the
same lot of seeds used for the phenotyping. The seeds experienced 1 year of after-ripening
before they were used for this experiment. For each genotype and parental environment, we
had three biological replicates that consist of a single seed per replicate. For each library,
mRNA was isolated from a single seed using Dynabeads mRNA DIRECT Kit from
Invitrogen (Product # 610.2, Grand Island, NY) and fragmented using Ambion mRNA
Fragmentation Kit (Product #AM8740, Grand Island, NY), followed by cDNA synthesis
using random hexamer primers. Double stranded cDNA fragments were blunt-end repaired
using Epicentre End Repair (Product #ER81050, Madison, WI) and added a single A to the
blunt end using Klenow Fragment 3’-5’ exo-nuclease (NEB Product #M0212L, Ipswich,
MA). Ilumina adaptors were ligated to the cDNA fragments using the Epicentre Fast-Link
95
DNA Ligation Kit (Product #LK6201H, Madison, WI). Fragments between 200-400bp
were size selected by agarose gel and samples were indexed and enriched. The libraries were
indexed using 12 indices and sequenced on two lanes using the Illumina GAIIx, which
generated 76 bp single-end reads.
4.3.3 Mapping and normalization of sequencing reads
To identify which transcripts are stored in seed tissues across the experiment, all reads from
all samples were pooled and mapped to the M. truncatula 3.5.1 genome sequence. We used
Tophat to map reads to the genome and generated full-length fragments (Roberts et al., 2012).
These fragments were assembled using Cufflinks (Roberts et al., 2012) to identify the
corresponding gene annotations. To identify differential expression patterns, the sequenced
reads from each sample were then analyzed independently using Cufflinks to generate counts
and coverage for seed-expressed genes and their isoforms. All samples were normalized
using the TMM protocol implemented using the edgeR package in R (Robertson et al., 2010),
which takes into account differences in overall RNA populations across biological samples.
4.3.4 Offspring phenotype data analysis
The influence of genotype (G), parental environment (PE), and offspring environment (OE)
on offspring traits (i.e. germination timing, timing of unifoliate and first trifoliate
development, leaf size, leaf number) was analyzed using the ANOVA package in R. We
considered all possible interactions between the main effects, and included seed weight as a
covariate. Data were Natural logarithm and square root transformation prior to analyses to
satisfy the model’s assumptions of normality and homoscedasticity. Genetic variation in
96
transgenerational plasticity would be indicated by a significant GxPE or GxPExOE
interaction with Bonferroni correction.
4.3.5 Seed transcriptome data analysis
For the seed transcriptome data, we used a negative binomial generalized linear model to
analyze the contributions of G, PE, and their interaction on the expression level of 4,358
expressed genes. Within a genotype-treatment combination, a gene was classified as
expressed when at least two out of three biological replicates had FPKM values greater than
1. We analyzed the contributions of the G and PE on the level of gene expression only for
genes that were expressed in all four genotypes. We used MASS (http://stat.ethz.ch/R-
manual/R-patched/library/MASS/html/glm.nb.html), to run the negative binomial
generalized linear model and analyzed the contribution of PE for the individual genotypes
using a false-discovery rate threshold of FDR < 0.05. The seed transcriptome was analyzed
for overrepresentation of biological processes terms using GOstat [30] and GO annotations
were obtained from the Noble Foundation (http://mtgea.noble.org/v3/). The data set
supporting the results of this article is available in the NCBI Short Read Archive repository
under ID SRP012122.
4.3.6 Network and functional analysis
Seed coexpression network topology file was downloaded from the SeedNet database
(bree.cs.nott.ac.uk/arabidopsis/ ) and visualized using Cytoscape version 2.8.3 (Shannon et al.,
2003). Cytoscape was used for network visualization and functional analysis of the seed
transcriptome (Shannon et al., 2003). The jActiveModules plugin was used to identify the
interaction modules referenced from a model of genome-wide transcriptional interaction
97
network derived from publically available microarray expression data of Arabidopsis mature
seeds from the SeedNet database (Bassel et al., 2008). Putative sub-networks with aggregate Z-
scores greater than 3.0 are generally considered significant and sub-networks were chosen
for analysis according to highest ranked Z-scores. Because jActiveModules relies on random
sampling, we ran several iterations of the data set to ensure the reproducibility of the
identified modules. To extract meaningful molecular associations from the complex sub-
networks identified, transcriptional interaction modules were further partitioned into tightly
linked coexpression clusters using the MCODE plugin (Bader and Hogue, 2003). MCODE
cluster scores greater than 2.0 were considered meaningful (Kirkpatrick and Lande, 1989; Cline
et al., 2007; Hoyle and Ezard, 2012) and clusters analyzed for this study were selected by highest
ranked. Since we assume that genes in a module are involved in the same biological process,
the predicted modules and clusters were validated by determining if the interacting nodes
(genes) are enriched for any Gene Ontology (GO) biological processes using the BiNGO
plugin. The hypergeometric test along with the Benjamini and Hochberg false discovery rate
correction for multiple testing with a p-value threshold of 0.001 were used to identify
significant overrepresented GO terms (Maere et al., 2005; Marshall, 2008; Donohue, 2009b; Dyer
et al., 2010).
4.4 Results
Stored seed transcripts play a critical role in germination (Rajjou, 2004; Mousseau et al., 2009;
Hoyle and Ezard, 2012) and could potentially mediate transgenerational plastic germination
responses; therefore, we sequenced the transcriptome of dry, mature seeds originating from
parental plants exposed to saline and non-saline conditions. A subset of the mature dry seeds
98
was used for transcriptome sequencing of stored seed transcripts, while the other subset of
seeds were used to quantify seed size and germination timing in saline and non-saline
offspring conditions. Because M. truncatula seeds are self-fertilized, the parental
environmental effect incorporates both maternal and paternal effects. We note that these
seeds were dry and hence the transcript accumulation is not influenced by seed germination
but rather reflects the deposition of storage transcripts during seed maturation.
4.4.1 Sequencing stored seed transcriptome
We report an average of 1.5 million mapped 76 bp reads for each library, resulting in ~6X
coverage of sequenced genes. Though our results are conservative, we identified 9,281 genes
expressed in seeds, which is more than the 2,759 genes identified in M. truncatula mature
seeds (Kirkpatrick and Lande, 1989; Benedito et al., 2008; Storm and Lima, 2010; KOZAK and
BOUGHMAN, 2012; Hoyle and Ezard, 2012) and less than the ~12,000 genes expressed in
Arabidopsis seeds (Agrawal, 2002; Nakabayashi et al., 2005; Marshall, 2008; Donohue, 2009b; Dyer et
al., 2010) and the 17,000 genes expressed in rice (Van Dam and Baldwin, 2001; Howell et al.,
2008; Reed et al., 2010). Gene ontology (GO) enrichment analysis revealed significant
enrichment of biological processes (Table 4.1), some of which are involved in gene
expression (GO:0010467), RNA metabolic processes/RNA processing (GO:0016070,
GO:0006396), translation (GO:0006412), response to osmotic stress (GO:0006970),
chromatin modification (GO:0016568), RNA splicing (GO:0008380) and miRNA-mediated
gene silencing (GO:0035196). Furthermore, M. truncatula forms symbiotic relationships with
nitrogen-fixing rhizobia and we find significantly under-enrichment of genes involved in
nodulation (GO:0009877) and symbiosis (GO:0044419) processes in the seed transcriptome,
suggesting that the symbiosis pathway is independent of the germination pathway.
99
Gene Ontology Biological Process P-value
GO:0044260 Cellular macromolecule metabolic process 7.38e-34
GO:0010467 Gene expression 6.97e-29
GO:0044265 Cellular macromolecule catabolic process 6.22e-28
GO:0044238 Primary metabolic process 7.29e-27
GO:0006508 Proteolysis 2.63e-21
GO:0015031 Protein transport 3.77e-21
GO:0019941 Modification-dependent protein catabolic process 4.57e-20
GO:0006412 Translation 4.05e-13
GO:0022613 Ribonucleoprotein complex biogenesis and assembly 3.37e-11
GO:0016070 RNA metabolic process 6.25e-11
GO:0006970 Response to osmotic stress 3.57e-08
GO:0006396 RNA processing 6.23e-07
GO:0009657 Plastid organization and biogenesis 1.28e-06
GO:0006807 Nitrogen compound metabolic process 1.84e-06
GO:0006325 Establishment and maintenance of chromatin architecture 5.38e-06
GO:0007275 Multicellular organismal development 3.05e-05
GO:0009628 Response to abiotic stimulus 3.70e-05
GO:0006464 Protein modification process 6.28e-05
GO:0044419 Interspecies interaction between organisms -7.45e-05
GO:0009791 Post-embryonic development 7.62e-05
GO:0051276 Chromosome organization and biogenesis 7.72e-05
GO:0009877 Nodulation -9.72e-05
GO:0044403 Symbiosis encompassing mutualism through parasitism -1.09e-04
GO:0022414 Reproductive process 3.76e-04
GO:0016568 Chromatin modification 6.86e-04
GO:0008380 RNA splicing 6.86e-04
GO:0032774 RNA biosynthetic process 8.97e-04
GO:0006351 Transcription DNA-dependent 8.97e-04
GO:0019725 Cellular homeostasis 1.67e-04
GO:0035196 miRNA-mediated gene silencing production of miRNAs 1.72e-03
GO:0000375 RNA splicing via transesterification reactions 6.50e-03
GO:0042221 Response to chemical stimulus 8.32e-03
GO:0044271 Nitrogen compound biosynthetic process 8.99e-03
GO:0016071 mRNA metabolic process 0.0139
GO:0006457 Protein folding 0.0143
GO:0031047 RNA-mediated gene silencing 0.0160
GO:0006355 Regulation of transcription DNA-dependent 0.0202
GO:0009887 Organ morphogenesis 0.0207
GO:0009266 Response to temperature stimulus 0.0278
GO:0006338 Chromatin remodeling 0.0278
GO:0043487 Regulation of RNA stability 0.0283
GO:0006402 mRNA catabolic process 0.0283
GO:0006399 tRNA metabolic process 0.0304
GO:0006413 Translational initiation 0.0353
Table 4.1 Complete list of significant gene ontology terms for total genes expressed in the seed
transcriptome enriched in biological pathways.
100
4.4.2 Stored seed transcripts are annotated to be involved in germination
and dormancy processes
We then asked whether the transcripts detected in M. truncatula overlapped with those
known to be involved in germination in A. thaliana. Using the gene annotations of A. thaliana
homologues, we compared the genes expressed in our seed transcriptome with genes
characterized in the Arabidopsis seed coexpression network (Galloway and Etterson, 2007;
Bassel et al., 2008; Sultan et al., 2009). We found that 58% of genes expressed in M. truncatula
seeds (5359/9281) are involved in regulating seed germination and dormancy processes,
accounting for 62% of genes represented in the Arabidopsis seed network (5359/8621). This
overlap is disproportionately higher than expected by chance (Fisher-test, p-value= 0.0025),
indicating that we captured seed transcripts that are involved in regulating germination and
dormancy processes.
4.4.3 Genotype-dependent parental environmental effects on stored seed transcripts
To examine the genetic basis of transgenerational plastic responses to salinity stress, we
exposed four parental genotypes to saline and non-saline environments and collected seeds
from each genotype per parental environment. To quantify the genetic differences in the
expression of the transcriptome of dry, mature seeds, we tested genotype (G) and parental
environmental effects (PE) on the expression of stored seed transcripts. A GxPE interaction
indicates genotypic differences in parental environmental influence on the expression of
stored seed transcripts. Correcting for multiple testing, we found 1,362 genes that respond to
the parental environment in a genotype dependent way (GxPE; all FDR<0.05), along with
1,500 genes that vary in expression across genotypes (G) irrespective of the parental
101
environment, and 471 genes that are responsive to the parental environment (PE) across all
genotypes.
Because we detected genetic variation in parental environmental effects (GxPE) on the
expression levels of stored seed transcripts, we analyzed TN1.13, TN1.15, TN7.22 and
TN8.22 separately, and detected 1195, 327, 125 and 82 genes differentially expressed
between parental environments, respectively. Furthermore, there is minimal overlap of PE
responsive transcripts between the genotypes and no genes shared among all four genotypes,
suggesting genotypic differences in salt stress response mechanisms. Genotypes TN1.13 and
TN1.15 possessed significantly more PE responsive transcripts than genotypes TN7.22 and
TN8.22 (Fisher Test, p-value=3.227e-08).
4.4.4 Genotype-dependent transgenerational plasticity on germination
behavior
To quantify the genetic basis of transgenerational plastic germination behavior, we tested G,
PE, OE and all possible interactions on germination timing. A significant OE term indicates
germination plasticity in response to the offspring environment, while a significant PE term
indicates parental environmental influence on offspring’s germination response. A significant
PExOE indicate transgenerational plasticity or parental environmental effect that is
dependent on offspring environment. A GxPExOE interaction indicates genetic variation of
trangenerational plasticity on offspring germination response that is dependent on the
relationship of the parental and offspring environment. Given that we detected genotypic
effects in the expression levels of stored seed transcripts and that previous experiments with
these genotypes found phenotypic differences in transgenerational plasticity (Lacey and
102
Herr, 2000; Galloway and Etterson, 2007; Mousseau et al., 2009; Hoyle and Ezard, 2012;
Moriuchi et al., 2014), we predict genetic variation in transgenerational plasticity of
germination timing. Indeed, we find genotype-dependent transgenerational plasticity (GxPE)
on germination timing of the same seed lot in which we measured stored seed transcript
differences.
We detected a significant three-way interaction (GxPExOE) for germination timing that
reflects genotype dependent transgenerational plasticity on germination behavior, timing of
unifoliate and first trifoliate development (F
(3,138)
=7.75, P < 0.00001;
Table 4.2). Consistent
with the expectations from preliminary germination data, TN1.13 and TN1.15 offspring
were the only genotypes that exhibited germination timing that depended on the interaction
between the parental and offspring environment (PExOE, F
(1,32)
=9.366, F
(1,34)
=19.964;
Bonferroni correction: P < 0.00125; Table 4.3).
103
Age at
germination
Age at
unifoliate dev.
Age at
trifoliate dev.
G 0.3325 0.6322 0.9837
PE 0.2479 0.3300 0.3819
OE 0.0097 0.0008 0.0236
Seed weight 0.9494 0.3253 0.1751
G x PE 0.2027 0.3209 0.4080
G x OE 0.0679 0.6124 0.6907
PE x OE 0.0006 0.0030 0.0146
G x PE x OE 0.0002 0.0172 0.0256
Adjusted-R
2
0.54 0.55 0.42
Table 4.2 ANOVA P-values for offspring traits. Offspring traits are explained by genotype (G),
parental environment (PE), offspring environment (OE) and the interaction terms, with seed weight
as a covariate for age of germination, age of unifoliate and first trifoliate development.
TN1.13 TN1.15 TN7.22 TN8.22
PE 0.297 1.377 0.132 0.062
OE 46.060*** 32.638*** 48.364*** 32.042***
PExOE 9.366** 19.964*** 0.297 3.107
Seed mass 0.422 0.751 0.325 0.121
Df/Res. Df 1/32 1/34 1/34 1/35
*P < 0.0125 **P < 0.001 ***P < 0.0001
Table 4.3 ANOVA F-values for offspring germination timing. Germination timing explained by the
parental environment (PE), offspring environment (OE) and all possible interactions with seed mass
as a covariate for each genotype. Bold values indicate significant effects. Df: degrees of freedom Res.
Df: residual degrees of freedom. Bonferroni correction for multiple comparisons P < 0.0125.
These genotypes only show significant differences in germination responses to the offspring
environment when the parental environment was saline, but no such differences were seen
when the parental environment was non-saline (Figure 4.1 a-b, red asterisk). In contrast,
TN7.22 and TN8.22 displayed no significant transgenerational plasticity on germination
104
timing (TN7.22: F
(1,34)
=0.297, P > 0.0125; TN8.22: F
(1,35)
=3.107, P > 0.0125; Figure 4.1 c-d,
Table 4.3), but germination differences in response to salinity were primarily driven by the
offspring environment (Figure 4.1 c-d, Table 4.3).
Figure 4.1 Norms of reaction plots for germination timing in response to parental and offspring
environment. Blue and red lines correspond to 0mM and 100mM NaCl parental environment,
respectively. The black asterisk indicates significantly different means with respect to parental
environments, and red asterisk indicates significantly different means with respect to offspring
environment (Mann-Whitney Test, P < 0.05).
In summary, these results show genotype-dependent transgenerational plasticity of offspring
germination timing. Among the genotypes that express transgenerational plastic germination
response (TN1.13 and TN1.15), offspring plasticity was dependent on the parental
environment: only parental plants that experienced saline conditions produced offspring
with plastic germination responses to offspring environment (Figure 4.1). Furthermore, we
105
find that the offspring environment has a larger effect on offspring germination timing than
transgenerational effects (Table 4.3).
4.4.5 Genotype-dependent parental environmental effects on seed size
and the absence of seed size effects on germination behavior
We test both genetic and parental environmental effects on seed size variation and examine
the relationship of seed size on germination behavior. Seed size differences between saline
and non-saline parental conditions varied among genotypes: the parental environment
significantly influenced seed size in genotypes TN7.22 and TN8.22, but had no influence on
seed size in genotype TN1.13 and TN1.15 (Figure 4.2). TN7.22 and TN8.22 parental plants
exposed to saline conditions produced larger seeds relative to non-saline conditions, while
TN1.13 and TN1.15 parental plants produced seeds that did not differ in size relative to
treatment conditions (Figure 4.2). Because parental environmental effects on seed size is
often correlated with germination timing (Crawley and Nachapong, 1985; Biere, 1991;
Schmid and Dolt, 1994; Galloway, 2001; Galloway and Etterson, 2007; Cendán et al., 2011;
Castro et al., 2013), we tested the effects of seed size on germination timing by including it as
a covariate in the analysis of variance. We found no relationship between seed mass and
germination timing across all four genotypes (Table 4.2 & 4.3).
106
Figure 4.2. Seed weight comparison of genotypes developing in 0mM and 100 mM parental NaCl
conditions.
4.4.6 Salt responsive mature seed transcripts are involved in dormancy
and ABA-related processes
Since an organism’s response to environmental cues are not guided by the expression
changes of just one gene but rather a network of interacting genes, finding differentially
expressed transcripts may not tell us much about how these transcripts work together to
influence the expression of a trait. Coexpression networks devised from correlated gene
expression is a powerful approach to find functional relationships between changes in gene
2.0 2.5 3.0 3.5 4.0 4.5 5.0
Parental Environment
Seed Weight (mg)
0mM 100mM
2.0 2.5 3.0 3.5 4.0 4.5 5.0
Legend
TN1.13
TN1.15
TN7.22
TN8.22
107
expression and phenotypic response. Coexpression networks work on a basic principle that
genes involved in a biological pathway are co-regulated; thus, coexpressed genes are more
likely to function in the same biochemical or developmental pathways (Bradshaw, 1965;
Schlichting, 1986; Hughes et al., 2000; Aoki et al., 2007; Saito et al., 2008; Usadel et al., 2009;
Mitra et al., 2013).
To identify pathways potentially involved in mediating transgenerational plastic germination
responses, we used Cytoscape to assess and visualize the coexpression networks of genes
responsive to parental exposure to salinity stress within the context of the Arabidopsis seed
coexpression network (Levin and Kerster, 1974; Cendán et al., 2011). This genome-wide co-
expression network describes transcriptional interactions of dormant and germinating seeds
that were derived from expression meta-data generated exclusively from mature Arabidopsis
seeds (Figure 4.3). This network consists of 8,261 nodes (genes) that is comprised of distinct
regions of clustered interaction enriched in transcripts identified in microarray data (Figure
4.3 a). Outlined in yellow are defined regions in the network: region 1 represents clusters of
genes associated with nongermination/dormancy; region 2 represents a transition from
nongerminating/dormancy to germinating states; region 3 is associated with germination.
108
Figure 4.3 SeedNet coexpression networks topology. (a.) The Arabidopsis germination coexpression
network. Outlined in yellow are defined regions in the network: region 1 represents clusters of genes
associated with nongermination/dormancy; region 2 represents a transition from
nongerminating/dormancy to germinating states; region 3 is associated with germination.
Subnetworks generated from salt responsive transripts expressed in genotypes (b.) TN1.13 and (c.)
TN1.15. (d.) Subnetwork of overlapping salt responsive genes of TN1.13 and TN1.15. The red and
blue nodes represent transcripts significantly associated with the upregulation of dormancy and
germination, respectively. Yellow nodes in the subnetworks of (b.) and (c.) highlight genes of highly
clustered modules.
109
We queried the genes responsive to salinity stress in the Arabidopsis seed co-expression
network among the four genotypes and found sub-networks associated with distinct regions
of the seed germination network (Figure 4.3 a-d). No significant sub-networks were found
for TN7.22 and TN8.22 saline responsive transcripts, which parallels the lack of
transgenerational plasticity in germination behavior of these two genotypes. In contrast, we
identified significant sub-networks for both TN1.13 and TN1.15 that are associated with
regions that correspond to seed germination in the Arabidopsis seed network (Figure 4.3 b-c,
region 3). To extract more meaningful co-regulated gene interactions from the complex sub-
networks (Figure 4.3b and 4.3c), we implemented MCODE to identify modules of tightly
clustered co-expressed genes that represent groups of genes that are potentially functioning
in the same biological pathway. We identified modules within the salt responsive sub-
networks (Figure 4.3b-c, yellow nodes) and found clusters significantly enriched for GO
terms associated with RNA metabolism, translation, and leaf development (Figure 4.4 a-b,
reference Appendix A for the complete list of genes corresponding to salt responsive
transcripts).
110
Figure 4.4 Discrete tightly clustered modules representing putative biological pathways. (a.)
MCODE cluster identified from TN1.13 subnetwork (Figure 3b) and TN1.15 subnetwork (Figure
3c). (b.) Functional interactions between genes associated with significant overrepresented GO terms.
Because TN1.13 and TN1.15 offspring exhibit similar patterns of parental environmental
effects on germination response to salinity (Figure 4.1), we queried the overlapping salt
responsive genes to identify the network relationship among these genes. Interestingly, the
sub-network derived from these overlapping salt responsive genes correspond to regions of
the seed network that are associated with seed dormancy (Figure 4.3d, region 1), supporting
the shared delayed germination responses to salinity we observed in these two genotypes.
Furthermore, among the shared salt responsive transcripts in TN1.13 and TN1.15 that are
involved in the seed dormancy network (Figure 4.3d), we find genes associated with ABA
up-regulation (Table 4.4) that account for over 30% of the dormancy related genes.
111
A. thaliana
genes
M. truncatula
orthologs
Annotation
At5g24670 Medtr8g103020 TRNA Adenosine Deaminase 3; involved in RNA editing
At5g62600 Medtr7g037000 MOS14; Nuclear importer of serine-arginine rich (SR) proteins;
involved in the regulation of splicing R genes
At4g35800 Medtr5g023020 NRPB1 DNA-directed RNA polymerase; DNA methylation, gene
silencing by RNA, RNA splicing
At2g44200 Medtr2g028200 CBF1-interacting co-repressor; Pre-mRNA splicing factor
At3g54280 Medtr4g034920 RGD3-Root Growth Defective 3; Chromatin remodeling complex
At5g13300 Medtr7g020860 SFC (SCARFACE); involved in response to auxin
At2g38410 Medtr7g072310 VHS domain-containing protein / GAT domain-containing protein;
involved in intracellular protein transport
At2g18700 Medtr4g129270 ATTPS11; Arabidopsis thaliana trehalose phosphatase/synthase 11;
involved in trehalose biosynthesis
At2g39340 Medtr3g073080 SAC3/GANP family protein
Table 4.4 Genes associated with ABA up-regulation within the dormancy coexpression sub-
network. Arabidopsis thaliana genes and M. trucatula orthologs of overlapping TN1.13 and TN1.15 salt
responsive genes that are involved in up-regulating ABA in the dormancy pathway.
4.5 Discussion
The goal of this study is to test for genetic variation in transgenerational plastic germination
behavior of four natural M. truncatula genotypes and identify stored seed transcripts that are
in involved in salt response. This is the initial step to identify potential genes and molecular
pathways that mediate germination response to salinity stress. We demonstrate genetic
variation in transgenerational plasticity of germination timing upon parental exposure to
salinity stress. The parental environment influences seed size in some genotypes, but there is
no overall relationship between seed size and germination behavior. This suggests that
112
parental environmental signals other than resource investment in seed size are important
influences on germination timing. Our study shows genotype-dependent parental
environmental effects on the expression level of stored seed transcripts. Among the
transcripts responsive to parental exposure to salinity stress, we identified genes associated
with seed dormancy pathway that may facilitate delayed germination response in saline
conditions. This may reflect an adaptive salt avoidance strategy as seen in other pasture
legumes (Mousseau et al., 2009; Plaistow and Benton, 2009; Nichols et al., 2009; Boots and
Roberts, 2012; Hoyle and Ezard, 2012; Ezard et al., 2014). Overall, our results suggest that
transgenerational plasticity can play an ongoing role in adaptation to saline habitats in M.
truncatula and identifies molecular pathways that may underlie the modulation of germination
behavior under salt stress.
Parental environmental effects on seed size may reflect differential parental resource
investment in offspring, a transgenerational mechanism that influences offspring
development and performance. Previous studies have suggested that parental environmental
effects on seed size influences germination timing (Biere, 1991; Lacey et al., 1997; Vange et
al., 2004; Galloway and Etterson, 2007). In Plantago lanceolata, seed size variation induced by
the parental environment during seed maturation is primarily driven by the seed coat rather
than the embryo and endosperm(Roach and Wulff, 1987; Lacey, 1996; Rossiter, 1996; Lacey
et al., 1997; Lacey and Herr, 2000): heavier seeds tend to have heavier or thicker seed coats
that delay germination response. In this study, where the seed coat was disrupted to focus on
physiological dormancy, we find that seed size did not correlate with germination timing in
any of the four genotypes. Instead we find that parental genotypes TN7.22 and TN8.22
exposed to salinity stress produced larger seeds that confer a growth advantage: plants
113
originating from large seeds tend to develop larger leaves and produce more leaves in
comparison to plants originating from smaller seeds (Table 4.5). These results suggest that
M. truncatula genotypes respond to salinity stress by increasing resource investment into
endosperm or embryo size rather than the seed coat. However, further studies in M.
truncatula addressing the partitioning of parental investment to seed traits in response to
environmental stress are necessary to understand how the parental environment shapes
offspring development and response through resource provisioning to seeds.
Seed size
TN1.13 TN1.15 TN7.22 TN8.22
Leaf size
0.05
[-0.28,0.38]
0.13
[-0.19,0.43]
0.61***
[0.36,0.78]
0.58***
[0.31,0.76]
Number of
leaves
-0.18
[-0.48,0.15]
-0.15
[-0.17,0.44]
0.49**
[0.21,0.70]
0.35*
[0.04,0.6]
*P < 0.05; **P < 0.001; ***P < 0.0001
Table 4.5 Pearson correlation coefficients for seed size in relation to leaf size and leaf counts for all
four genotypes. Bold coefficients are significant and asterisk corresponds significance level.
Early germination and increased parental investment into seed size confers a growth
advantage that may be indicative of salt tolerance rather than salt avoidance. Seeds from
TN7.22 and TN8.22 parental plants exposed to salinity tended to germinate earlier in saline
conditions irrespective of the offspring environment (Figures 4.1c and 4.1d), which could
potentially be an unfavorable response in saline habitats. However, we find that early
germination is correlated with higher growth potential in leaf size (Pearson correlation, r = -
0.49, P < 0.0001) and number of leaves (Pearson correlation, r = -0.32, P < 0.0001). A
114
previous study of Tunisian saline adapted genotypes showed that some genotypes exhibit a
level of salt tolerance during germination by modulating metabolic and physiological
processes to maintain ion balance in the root system (Biere, 1991; Donohue, 2009a; Cordeiro
et al., 2014) to improve water uptake important for photosynthesis and growth during salinity
stress. Overall, we detect parental environmental effects on offspring performance that does
not depend on the offspring, where some parental genotypes invest resources into seed size
to optimize offspring performance in the next generation.
In contrast, we find that some parental genotypes influence offspring germination and
seedling development to ensure seedling survival in saline conditions. Genotypes TN1.13
and TN1.15 displayed significant transgenerational plastic germination responses: offspring
delayed germination in saline conditions only when parental plants were exposed to salinity
stress. This may represent a viable salt avoidance mechanism to ensure seedling survival in
saline environments. Because saline habitats experience high levels of salinity early in the
rainy season and levels begin to dissipate with subsequent rain (Smith and Stoneman, 1970;
Eriksson, 1999; Chacón and Bustamante, 2001; Nichols et al., 2010), pasture legumes
adapted to these habitats have evolved delayed germination to avoid toxic levels of salinity
early in the growing season (Nichols et al., 2009; 2010; Donohue et al., 2010). In addition to
germination, these genotypes also show transgenerational plasticity in the timing of unifoliate
and first trifoliate development (Table 4.2, data not shown), suggesting that parental
environmental effects persist past seedling establishment and into early seedling
development. Further, there were no seed size effects on germination behavior and no
significant seed size differences between parental environments (Table 4.6, Figure 4.2),
115
suggesting that parental environmental signals other than resource investment in seed size
may play a role in modulating germination timing in these two genotypes.
Table 4.6 ANOVA on seed weight (g) differences between 0 mM and 100 mM parental
environment for each genotype. Bonferroni correction for multiple comparisons P < 0.0125.
Because stored seed transcripts play a pivotal role in seed germination (Crawley and
Nachapong, 1985; Schmid and Dolt, 1994; Rajjou, 2004; Moshatati and Gharineh, 2012), we
hypothesize that stored seed transcripts may represent a transgenerational mechanism some
parental genotypes exploit to influence offspring germination behavior. We detected
genotype-dependent parental environmental effects on the expression of stored seed
transcripts, with some involved in seed dormancy and germination pathways (Figure 4.3).
Previous studies have shown that the expression of specific stored seed transcripts
influenced by parental exposure to stress is correlated with the magnitude of seed dormancy
(Almoguera and Jordano, 1992; Jones et al., 1997; Kendall et al., 2011; Cendán et al., 2011). A
recent study in Arabidopsis demonstrated that cold stress induced seed transcripts in dry,
mature seeds were associated with genes that regulate seed dormancy and germination
timing (Williams and Harper, 1965; Donohue et al., 2010; Kendall et al., 2011). Because
TN1.13 and TN1.15 offspring exhibit similar patterns of parental environmental effects on
germination response to salinity (Figure 4.1), we queried the overlapping salt responsive
genes to identify the network relationship among these genes. Interestingly, we identified a
sub-network that corresponds to a region of the seed network associated with seed
F-value Coefficient P-value
TN1.13 3.85
(138)
3.85 0.06
TN1.15 6.01
(138)
6.01 0.02
TN7.22 254.5
(138)
254.51 2.2e-16
TN8.22 9.37
(1.38)
9.37 0.004
116
dormancy (Figure 4.3d, region 4.1), thus supporting the shared delayed germination
responses to salinity observed in these two genotypes.
A plant's ability to tolerate or adapt to salt stress often comes with a cost to vegetative
growth and reproduction due to both genetic and resource limitations (Van Dam and
Baldwin, 2001; Läuchli and Grattan, 2007; Kendall et al., 2011). M. trucatula plants adapted to
saline habitats are typically smaller and exhibit reduced reproductive output compared to
plants adapted to non-saline habitats (Baskin and Baskin, 1998; Russi et al., 2008; Bolingue et
al., 2010; Friesen et al., 2015). In our study, offspring exposed to saline conditions, on
average, produced fewer leaves than offspring growing in non-saline conditions (Wilcoxon
rank sum test, W = 16, P <0.05). Delayed germination is often correlated with reduced
performance and competitive ability (Rice and Dyer, 2001; Matthews and Hosseini, 2006;
Garcia et al., 2006; Taiz and Zeiger, 2010), while early seedling emergence is associated with
increased fitness (Roach and Wulff, 1987; Lacey et al., 1997; Verdú and Traveset, 2005).
However, if early germination results in seedling death, then we would expect selection to
favor delayed over early germination at the expense of reduced growth potential. Here we
find that delayed germination is correlated with smaller leaf size (Pearson correlation, r = -
0.32, P < 0.0001) and reduced number of leaves (Pearson correlation, r = -0.62, P <0.00001).
This may be due to tradeoffs between survival and performance in stressful environments
(Rees, 1994; Bolingue et al., 2010) Here we find that some genotypes may exhibit this
tradeoff in germination timing and this is likely due to genotypic differences in the degree of
salt tolerance and parental strategy to cope with salinity stress.
117
Despite the fitness cost of delayed germination, dormancy mechanisms have evolved as an
adaptive response to environmental uncertainty across many species of plants (Baskin and
Baskin, 1978; Venable and Brown, 1988; Bewley, 1997; Baskin and Baskin, 1998; Penfield
and Springthorpe, 2011). Germination timing is an important life history trait that
determines the environment of the developing plant; thus, influences seedling survival and
adaptive traits later in life (Roach and Wulff, 1987; Lacey et al., 1997; Donohue et al., 2010).
After-ripened non-dormant seeds can induce secondary dormancy when environmental
conditions are unfavorable (Bewley and Black, 1982; Baskin and Baskin, 1985; Cadman et al.,
2006; Finch-Savage et al., 2007). ABA is a pleiotropic plant hormone, playing key roles in a
variety of developmental pathways that include seed development and dormancy
(Almoguera and Jordano, 1992; Hoecker et al., 1995; Jones et al., 1997; Bassel et al., 2008;
Penfield et al., 2010; Bassel et al., 2011; Nakashima and Yamaguchi-Shinozaki, 2013), in
addition to adaptive stress responses to environmental perturbations in plants (Finkelstein
and Gibson, 2002; Rajjou, 2004; Cutler et al., 2010; Hubbard et al., 2010). Numerous studies
have observed ABA-mediated gene expression in response to drought and salt stress (Jones
et al., 1997; Fujita et al., 2011). In our study, among the shared salt responsive transcripts in
TN1.13 and TN1.15, we find transcripts associated with ABA up-regulation (Table 4.4) and
these transcripts are involved in the seed dormancy network (Figure 4.3d). These ABA
related genes account for 30% of the dormancy related genes and they are functionally
characterized to be involved in post-transcriptional regulation (TRNA Adenosine Deaminase
3, MOS14, NRPBI, CBF1), epigenetic mechanisms (NRPB1, RGD3) and growth and
developmental processes (SFC; Table 4.4; Figure 4.3d). This suggests a dynamic crosstalk
between ABA and gene regulatory pathways that potentially regulate transgenerational
plasticity in germination timing under parental salinity stress. Although phytohormone ABA
118
does not appear to be directly involved in secondary dormancy, transcription factors
regulated by ABA in dry, mature seeds may be involved in transitioning after-ripened non-
dormant seeds to secondary dormancy (Harada, 1997; Bewley, 1997; Ruuska et al., 2002;
Kendall et al., 2011). Although our results suggest that seeds may delay germination through
the induction of secondary dormancy, but it is unclear whether M. truncatula seeds undergo
secondary dormancy.
Recently, the transmission of epigenetic marks regulating gene expression (i.e., DNA
methylation) has emerged as a candidate mechanism mediating adaptive transgenerational
responses in plants (Almoguera and Jordano, 1992; Verhoeven et al., 2009; Kendall et al., 2011;
Luna et al., 2012). In fact, we found significant enrichment of seed transcripts associated with
chromatin remodeling and miRNA production (Table 4.1), which is consistent with the
results observed in Arabidopsis seed transcriptome (Nakabayashi et al., 2005; Badri et al., 2007;
Lazrek et al., 2009; Friesen et al., 2010; 2015). Although these processes have been documented
to mediate transgenerational effects through DNA methylation (Smith and Stoneman, 1970;
Nichols et al., 2008; Badyaev and Uller, 2009; Ho and Burggren, 2010), these mechanisms were not
implicated in our study as mediating salt response in the seed transcriptome. However, our
results do not discount the possibility that epigenetic mechanisms play a role in mediating
transgenerational effects, because our experiment was not designed to capture the influence
of these processes in response to salinity stress. Therefore, bisulfite and miRNA sequencing
in future experiments will be necessary to understand the relationship between epigenetic
mechanism and transgenerational plasticity in germination behavior.
119
Post-transcriptional processing of stored seed transcripts play a critical role in seed
germination (Rajjou, 2004; Nichols et al., 2009), and are likely involved in modulating
germination behavior when parental plants are exposed to salinity stress. In this study, we
find that alternative splicing and translation processes respond to parental exposure to
salinity stress (Figure 4.4), which might represent novel mechanisms mediating
transgenerational plasticity in seedling development. Because translation of stored seed
transcripts represent a critical step in seed germination (Crawford et al., 1989; Baskin and Baskin,
1998; Rajjou, 2004; Patanè and Gresta, 2006; Gallardo et al., 2006; Garcia et al., 2006), it is not
surprising that we find post-transcriptional regulatory networks involved in responding to
parental exposure to salinity stress. Genome-wide post-transcriptional regulation under
abiotic stress conditions have shown an important contribution of translational regulation of
stress responsive transcripts (Gallardo et al., 2006; Spriggs et al., 2010). The amount of specific
proteins translated is critical for biological pathways to function optimally in response to
environmental perturbations. Furthermore, different alternative splice variants of
transcription factors in response to environmental cues are involved in regulating seed
dormancy and germination in Arabidopsis (Gallardo et al., 2006; Finkelstein et al., 2008; Bolingue et
al., 2010; Penfield et al., 2010). Since TN1.13 and TN1.15 early offspring response to salinity
depends on both the parental and offspring environment, alternative splicing and
translational regulation may represent a viable transgenerational mechanism that can process
parental environmental cues to influence offspring response to salinity stress.
In this study, we have demonstrated that parental environmental signals can be transmitted
through the expression of stored seed transcripts or resources provisioned to the seed to
influence offspring response and development under salinity stress. Parental environmental
120
effects on seed dormancy permits temporal escape from unfavorable conditions early in the
germination season, while parental investment in larger seeds provide seedlings with
resources to improve performance under unfavorable conditions. Although the mature seed
transcriptome has been characterized in several plant species (Nakabayashi et al., 2005;
Benedito et al., 2008) including M. truncatula seeds (Badri et al., 2007; Benedito et al., 2008;
Lazrek et al., 2009; Friesen et al., 2010), none of these studies have linked the effect of stored
seed transcripts to transgenerational plastic germination behavior. Our study demonstrates a
potential for parental control over seed dormancy by influencing the expression of stored
seed transcripts and propose novel post-transcriptional mechanisms involved in germination
under salinity stress. Furthermore, the genotypic differences seen in parental environmental
effects on the expression of stored seed transcripts and offspring germination response
suggest that transgenerational plasticity of germination behavior can potentially evolve under
saline conditions.
4.6 References
Agrawal AA (2002). Herbivory and maternal effects: mechanisms and consequences of
transgenerational induced plant resistance. Ecology 83: 3408–3415.
Almoguera C, Jordano J (1992). Developmental and environmental concurrent expression of
sunflower dry-seed-stored low-molecular-weight heat-shock protein and Lea mRNAs.
Plant Mol Biol 19: 781–792.
Aoki K, Ogata Y, Shibata D (2007). Approaches for extracting practical information from
gene co-expression networks in plant biology. Plant and Cell Physiology 48: 381–390.
Bader GD, Hogue CWV (2003). An automated method for finding molecular complexes in
large protein interaction networks. BMC Bioinformatics 4: 2.
Badri M, Ilahi H, Huguet T, Aouani ME (2007). Quantitative and molecular genetic variation
in sympatric populations of Medicago laciniata and M. truncatula (Fabaceae):
relationships with eco-geographical factors. Genet Res 89: 107–122.
121
Badyaev AV, Uller T (2009). Parental effects in ecology and evolution: mechanisms,
processes and implications. Philosophical Transactions of the Royal Society B: Biological Sciences
364: 1169–1177.
Baskin CC, Baskin JM (1998). Seeds: ecology, biogeography, and evolution of dormancy and germination.
Academic Press.
Baskin JM, Baskin CC (1978). Temperature Requirements for Afterripening of Seeds of a
Winter Annual Induced into Secondary Dormancy by Low Winter Temperatures.
Bulletin of the Torrey Botanical Club 105: 104.
Baskin JM, Baskin CC (1985). The Annual Dormancy Cycle in Buried Weed Seeds: A
Continuum. BioScience 35: 492–498.
Bassel GW, Fung P, Chow TFF, Foong JA, Provart NJ, Cutler SR (2008). Elucidating the
Germination Transcriptional Program Using Small Molecules. PLANT PHYSIOLOGY
147: 143–155.
Bassel GW, Lan H, Glaab E, Gibbs DJ, Gerjets T, Krasnogor N, et al. (2011). Genome-wide
network model capturing seed germination reveals coordinated regulation of plant
cellular phase transitions. Proc Natl Acad Sci USA 108: 9709–9714.
Benedito VA, Torres-Jerez I, Murray JD, Andriankaja A, Allen S, Kakar K, et al. (2008). A
gene expression atlas of the model legume Medicago truncatula. The Plant Journal 55:
504–513.
Bewley JD (1997). Seed Germination and Dormancy. THE PLANT CELL ONLINE 9:
1055–1066.
Bewley JD, Black M (1982). Physiology and biochemistry of seeds in relation to germination. Vol. 2.
Viability, dormancy and environmental control. Springer-Verlag.
Biere A (1991). Parental effects in Lychnis flos-cuculi. I: Seed size, germination and seedling
performance in a controlled environment. Journal of Evolutionary Biology 4: 447–465.
Bolingue W, Ly Vu B, Leprince O, Buitink J (2010). Characterization of dormancy behaviour
in seeds of the model legume Medicago truncatula. Seed Sci Res 20: 97.
Boots M, Roberts KE (2012). Maternal effects in disease resistance: poor maternal
environment increases offspring resistance to an insect virus. Proceedings of the Royal Society
B: Biological Sciences 279: 4009–4014.
Bradshaw AD (1965). Evolutionary significance of phenotypic plasticity in plants. Advances in
genetics.
Cadman CSC, Toorop PE, Hilhorst HWM, Finch-Savage WE (2006). Gene expression
profiles of Arabidopsis Cvi seeds during dormancy cycling indicate a common
underlying dormancy control mechanism. The Plant Journal 46: 805–822.
122
Castro BM, Moriuchi KS, Friesen ML, Badri M, Nuzhdin SV, Strauss SY, et al. (2013).
Parental environments and interactions with conspecifics alter salinity tolerance of
offspring in the annual Medicago truncatula (P Klinkhamer, Ed.). J Ecol 101: 1281–1287.
Cendán C, Sampedro L, Zas R (2011). The maternal environment determines the timing of
germination in Pinus pinaster. Environmental and Experimental Botany: 1–7.
Chacón P, Bustamante RO (2001). The effects of seed size and pericarp on seedling
recruitment and biomass in Cryptocarya alba (Lauraceae) under two contrasting
moisture regimes - Springer. Plant Ecology 152: 137–144.
Cline MS, Smoot M, Cerami E, Kuchinsky A, Landys N, Workman C, et al. (2007).
Integration of biological networks and gene expression data using Cytoscape. Nat Protoc
2: 2366–2382.
Cordeiro MA, Moriuchi KS, Fotinos TD, Miller KE, Nuzhdin SV, Wettberg Von EJ, et al.
(2014). Population Differentiation for Germination and Early Seedling Root Growth
Traits Under Saline Conditions in the Annual Legume Medicago Truncatula (Fabaceae).
Am J Bot 101: 488–498.
Crawford EJ, Lake A, Boyce KG (1989). ADVANCES IN AGRONOMY - Google Books.
Advances in Agronomy.
Crawley MJ, Nachapong M (1985). JSTOR: Journal of Ecology, Vol. 73, No. 1 (Mar., 1985),
pp. 255-261. The Journal of Ecology.
Cutler SR, Rodriguez PL, Finkelstein RR, Abrams SR (2010). Abscisic acid: emergence of a
core signaling network. Annu Rev Plant Biol 61: 651–679.
Donohue K (2009a). Completing the cycle: maternal effects as the missing link in plant life
histories. Philosophical Transactions of the Royal Society B: Biological Sciences 364: 1059–1074.
Donohue K (2009b). Completing the cycle: maternal effects as the missing link in plant life
histories. Philosophical Transactions of the Royal Society B: Biological Sciences 364: 1059.
Donohue K, Rubio de Casas R, Burghardt L, Kovach K, Willis CG (2010). Germination,
Postgermination Adaptation, and Species Ecological Ranges. Annu Rev Ecol Evol Syst 41:
293–319.
Dyer AR, Brown CS, Espeland EK, McKay JK, Meimberg H, Rice KJ (2010). SYNTHESIS:
The role of adaptive trans-generational plasticity in biological invasions of plants.
Evolutionary Applications 3: 179–192.
Eriksson O (1999). Seed size variation and its effect on germination and seedling
performance in the clonal herb Convallaria majalis. Acta Oecologica 20: 61–66.
Ezard THG, Prizak R, Hoyle RB (2014). The fitness costs of adaptation via phenotypic
plasticity and maternal effects (D Marshall, Ed.). Funct Ecology 28: 693–701.
123
Finch-Savage WE, Cadman CSC, Toorop PE, Lynn JR, Hilhorst HWM (2007). Seed
dormancy release in Arabidopsis Cvi by dry after-ripening, low temperature, nitrate and
light shows common quantitative patterns of gene expression directed by
environmentally specific sensing - Finch-Savage - 2007 - The Plant Journal - Wiley
Online Library. Plant J 51: 60–78.
Finkelstein RR, Gibson SI (2002). ABA and sugar interactions regulating development:
cross-talk or voices in a crowd? Current Opinion in Plant Biology.
Finkelstein R, Reeves W, Ariizumi T, Steber C (2008). Molecular Aspects of Seed
Dormancy*. Annu Rev Plant Biol 59: 387–415.
Friesen ML, Cordeiro MA, Penmetsa RV, Badri M, Huguet T, Aouani ME, et al. (2010).
Population genomic analysis of Tunisian Medicago truncatula reveals candidates for
local adaptation. Plant J 63: 623–635.
Friesen ML, Wettberg EJB, Badri M, Moriuchi KS, Barhoumi F, Chang PL, et al. (2015). The
Ecological and Genomic Basis of Salinity Adaptation in Tunisian Medicago truncatula.
BMC Genomics in press: 1–47.
Fujita Y, Fujita M, Shinozaki K, Yamaguchi-Shinozaki K (2011). ABA-mediated
transcriptional regulation in response to osmotic stress in plants. J Plant Res 124: 509–
525.
Gallardo G, Lesignor C, Darmency M, Burstin J, Thompson R, Rochat C, et al. (2006). M.
truncatula Handbook: Seed Biology of M. truncatula. : 1–23.
Galloway LF (2001). Parental environmental effects on life history in the herbaceous plant
Campanula americana. Ecology 82: 2781–2789.
Galloway LF, Etterson JR (2007). Transgenerational Plasticity Is Adaptive in the Wild. Science
318: 1134–1136.
Garcia J, Barker DG, Journet E-P (2006). Medicago truncatula handbook: seed storage and
germination. : 1–9.
Harada J (1997). Seed Maturation and Control of Germination (B Larkins and I Vasil, Eds.).
4: 545–592.
Ho DH, Burggren WW (2010). Epigenetics and transgenerational transfer: a physiological
perspective. J Exp Biol 213: 3–16.
Hoecker U, Vasil IK, McCarty DR (1995). Integrated control of seed maturation and
germination programs by activator and repressor functions of Viviparous-1 of maize.
Genes & Development 9: 2459–2469.
Howell KA, Narsai R, Carroll A, Ivanova A, Lohse M, Usadel B, et al. (2008). Mapping
Metabolic and Transcript Temporal Switches during Germination in Rice Highlights
Specific Transcription Factors and the Role of RNA Instability in the Germination
124
Process. PLANT PHYSIOLOGY 149: 961–980.
Hoyle RB, Ezard THG (2012). The benefits of maternal effects in novel and in stable
environments. J R Soc Interface 9: 2403–2413.
Hubbard KE, Nishimura N, Hitomi K, Getzoff ED, Schroeder JI (2010). Early abscisic acid
signal transduction mechanisms: newly discovered components and newly emerging
questions. Genes & Development 24: 1695–1708.
Hughes TR, Marton MJ, Jones AR, Roberts CJ, Stoughton R, Armour CD, et al. (2000).
Functional discovery via a compendium of expression profiles. Cell 102: 109–126.
Jones HD, Peters NC, Holdsworth MJ (1997). Genotype and environment interact to
control dormancy and differential expression of the VIVIPAROUS 1 homologue in
embryos of Avena fatua. Plant J 12: 911–920.
Kendall SL, Hellwege A, Marriot P, Whalley C, Graham IA, Penfield S (2011). Induction of
Dormancy in Arabidopsis Summer Annuals Requires Parallel Regulation of DOG1 and
Hormone Metabolism by Low Temperature and CBF Transcription Factors. The Plant
Cell 23: 2568–2580.
Kirkpatrick M, Lande R (1989). JSTOR: Evolution, Vol. 43, No. 3 (May, 1989), pp. 485-503.
Evolution.
KOZAK GM, BOUGHMAN JW (2012). Plastic responses to parents and predators lead to
divergent shoaling behaviour in sticklebacks. Journal of Evolutionary Biology 25: 759–769.
Lacey EP (1996). Parental effects in Plantago lanceolata L. I.: A growth chamber experiment
to examine pre- and postzygotic temperature effects. Evolution 50: 865–878.
Lacey EP, Herr D (2000). Parental effects in Plantago lanceolata L. III. Measuring parental
temperature effects in the field. Evolution 54: 1207–1217.
Lacey E, Smith S, Case A (1997). Parental effects on seed mass: seed coat but not
embryo/endosperm effects. Am J Bot 84: 1617–1617.
Lazrek F, Roussel V, Ronfort J, Cardinet G, Chardon F, Aouani ME, et al. (2009). The use of
neutral and non-neutral SSRs to analyse the genetic structure of a Tunisian collection of
Medicago truncatula lines and to reveal associations with eco-environmental variables.
Genetica 135: 391–402.
Läuchli A, Grattan SR (2007). Plant Growth And Development Under Salinity Stress. In: …
in molecular breeding toward drought and salt …, Springer Netherlands: Dordrecht, pp 1–32.
Levin D, Kerster H (1974). Gene Flow in Seed Plants (T Dobzhansky, M Hecht, and W
Steere, Eds.). : 139–220.
Luna E, Bruce TJA, Roberts MR, Flors V, Ton J (2012). Next-Generation Systemic
Acquired Resistance. PLANT PHYSIOLOGY 158: 844–853.
125
Maere S, Heymans K, Kuiper M (2005). BiNGO: a Cytoscape plugin to assess
overrepresentation of gene ontology categories in biological networks. Bioinformatics 21:
3448–3449.
Marshall DJ (2008). Transgenerational plasticity in the sea: context-dependent maternal
effects across the life history. 89: 418–427.
Matthews S, Hosseini MK (2006). Mean germination time as an indicator of emergence
performance in soil of seed lots of maize (Zea mays). Seed Science and Technology 34: 339–
347.
Mitra K, Carvunis A-R, Ramesh SK, Ideker T (2013). Integrative approaches for finding
modular structure in biological networks. Nat Rev Genet 14: 719–732.
Moriuchi KS, Friesen ML, Cordeiro MA, Badri M, Vu WT, Main B, et al. (2014). The role of
parental and offspring environment on population responses of Medicago truncatula to soil salinity. in
review: American Journal of Botany.
Moshatati A, Gharineh MH (2012). Effect of grain weight on germination and seed vigor of
wheat. Intl J Agri Crop Sci.
Mousseau TA, Uller T, Wapstra E, Badyaev AV (2009). Evolution of maternal effects: past
and present. Philosophical Transactions of the Royal Society B: Biological Sciences 364: 1035–
1038.
Nakabayashi K, Okamoto M, Koshiba T, Kamiya Y, Nambara E (2005). Genome-wide
profiling of stored mRNA in Arabidopsis thaliana seed germination: epigenetic and
genetic regulation of transcription in seed. The Plant Journal 41: 697–709.
Nakashima K, Yamaguchi-Shinozaki K (2013). ABA signaling in stress-response and seed
development. Plant Cell Rep 32: 959–970.
Nichols PGH, Craig AD, Rogers ME, Albertsen TO, Miller SM, McClements DR, et al.
(2008). Production and persistence of annual pasture legumes at five saline sites in
southern Australia. Aust J Exp Agric 48: 518.
Nichols PGH, Malik AI, Stockdale M, Colmer TD (2009). Salt tolerance and avoidance
mechanisms at germination of annual pasture legumes: importance for adaptation to
saline environments. Plant Soil 315: 241–255.
Nichols P, Barrett-Lennard E, Bennett S (2010). Pasture legumes and grasses for saltland. :
1–4.
Patanè C, Gresta F (2006). Germination of Astragalus hamosus and Medicago orbicularis as
affected by seed-coat dormancy breaking techniques. Journal of Arid Environments 67:
165–173.
Penfield S, Springthorpe V (2011). Understanding chilling responses in Arabidopsis seeds
and their contribution to life history. Philosophical Transactions of the Royal Society B:
126
Biological Sciences 367: 291–297.
Penfield S, Josse E-M, Halliday KJ (2010). A role for an alternative splice variant of PIF6 in
the control of Arabidopsis primary seed dormancy. Plant Mol Biol 73: 89–95.
Plaistow SJ, Benton TG (2009). The influence of context-dependent maternal effects on
population dynamics: an experimental test. Trends in Ecology & Evolution 364: 1049–1058.
Rajjou L (2004). The Effect of -Amanitin on the Arabidopsis Seed Proteome Highlights the
Distinct Roles of Stored and Neosynthesized mRNAs during Germination. PLANT
PHYSIOLOGY 134: 1598–1613.
Reed TE, Waples RS, Schindler DE, Hard JJ, Kinnison MT (2010). Phenotypic plasticity and
population viability: the importance of environmental predictability. Proceedings of the
Royal Society B: Biological Sciences 277: 3391–3400.
Rees M (1994). Delayed Germination of Seeds - a Look at the Effects of Adult Longevity,
the Timing of Reproduction, and Population Age/Stage Structure. American Naturalist
144: 43–64.
Rice KJ, Dyer AR (2001). Seed aging, delayed germination and reduced competitive ability in
Bromus tectorum. Plant Ecology 155: 237–243.
Roach DA, Wulff RD (1987). JSTOR: Annual Review of Ecology and Systematics, Vol. 18
(1987), pp. 209-235. Annual Review of Ecology and Systematics: 1–28.
Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, et al. (2012). Differential gene
and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks.
Nat Protoc 7: 562–578.
Robertson G, Schein J, Chiu R, Corbett R, Field M, Jackman SD, et al. (2010). De novo
assembly and analysis of RNA-seq data. Nat Meth 7: 909–912.
Rossiter MC (1996). JSTOR: Annual Review of Ecology and Systematics, Vol. 27 (1996), pp.
451-476. Annual Review of Ecology and Systematics.
Russi L, Cocks PS, Roberts EH (2008). Coat thickness and hard-seededness in some
Medicago and Trifolium species. Seed Sci Res 2: 243–249.
Ruuska SA, Girke T, Benning C, Ohlrogge JB (2002). Contrapuntal networks of gene
expression during Arabidopsis seed filling. The Plant Cell 14: 1191–1206.
Saito K, Hirai MY, Yonekura-Sakakibara K (2008). Decoding genes with coexpression
networks and metabolomics - 'majority report by precogs'. Trends in Plant Science 13: 36–
43.
Schlichting CD (1986). The evolution of phenotypic plasticity in plants. Annual Review of
Ecology and Systematics 17: 667–693.
127
Schmid B, Dolt C (1994). Effects of maternal and paternal environment and genotype on
offspring phenotype in Solidago altissima L. Evolution: 1525–1549.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. (2003). Cytoscape: a
software environment for integrated models of biomolecular interaction networks.
Genome Research 13: 2498–2504.
Smith ST, Stoneman TC (1970). Salt Movement in Bare Saline Soils.
Spriggs KA, Bushell M, Willis AE (2010). Translational Regulation of Gene Expression
during Conditions of Cell Stress. Molecular Cell 40: 228–237.
Storm JJ, Lima SL (2010). Mothers Forewarn Offspring about Predators: A
Transgenerational Maternal Effect on Behavior. The American Naturalist 175: 382–390.
Sultan SE, Barton K, Wilczek AM (2009). Contrasting patterns of transgenerational plasticity
in ecologically distinct congeners. Ecology 90: 1831–1839.
Taiz L, Zeiger E (2010). Plant Physiology.
Usadel B, Obayashi T, Mutwil M, Giorgi FM, Bassel GW, Tanimoto M, et al. (2009). Co-
expression tools for plant biology: opportunities for hypothesis generation and caveats.
Plant, Cell & Environment 32: 1633–1651.
Van Dam NM, Baldwin IT (2001). Competition mediates costs of jasmonate-induced
defences, nitrogen acquisition and transgenerational plasticity in Nicotiana attenuata -
Van Dam - 2001 - Functional Ecology - Wiley Online Library. Funct Ecology.
Vange V, Heuch I, Vandvik V (2004). Do seed mass and family affect germination and
juvenile performance in Knautia arvensis? A study using failure-time methods. Acta
Oecologica-International Journal of Ecology 25: 169–178.
Venable DL, Brown CR (1988). The Selective Interactions of Dispersal, Dormancy, and
Seed Size as Adaptations for Reducing Risk in Variable Environments. American
Naturalist 131: 360–384.
Verdú M, Traveset A (2005). EARLY EMERGENCE ENHANCES PLANT FITNESS: A
PHYLOGENETICALLY CONTROLLED META-ANALYSIS.
http://dxdoiorg/101890/04-1647.
Verhoeven KJF, Jansen JJ, Van Dijk PJ, Biere A (2009). Stress-induced DNA methylation
changes and their heritability in asexual dandelions. New Phytologist 185: 1108–1118.
Williams JT, Harper JL (1965). Seed Polymorphism and Germination. Weed Res 5: 141–150.
128
Appendix A
List of M. truncatula and Arabidopsis orthologs involved in TN1.13 and TN1.15
transcriptional salt response.
AT MTR Annotation Type
AT1G78580 Medtr8g087910
Medtr8g087930
ATTPS1 (TREHALOSE-6-PHOSPHATE SYNTHASE); transferase
transferring glycosyl groups
Overlap
AT2G18700 Medtr4g129270 ATTPS11 (Arabidopsis thaliana trehalose phosphatase/synthase 11);
transferase transferring glycosyl groups
Overlap
AT3G08510 Medtr3g069280
Medtr3g070720
Medtr5g071010
Medtr5g082620
ATPLC2 (PHOSPHOLIPASE C 2); phospholipase C Overlap
AT3G12940 Medtr2g069480 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT3G19895.1);
similar to unnamed protein product "Vitis vinifera" (GB:CAO23819.1)
Overlap
AT3G14050 Medtr8g020880 RSH2 (RELA-SPOT HOMOLOG); catalytic Overlap
AT3G26720 Medtr7g084040 glycosyl hydrolase family 38 protein Overlap
AT4G18880 Medtr1g106840
Medtr8g087540
AT-HSFA4A (Arabidopsis thaliana heat shock transcription factor A4A);
DNA binding / transcription factor
Overlap
AT4G35580 Medtr3g093040 no apical meristem (NAM) family protein Overlap
AT5G01670 Medtr7g070500 aldose reductase putative Overlap
AT5G24670 Medtr8g103020 hydrolase/ zinc ion binding Overlap
AT5G51130 Medtr8g102950 contains InterPro domain Methyltransferase type 12 (InterPro:IPR013217);
contains InterPro domain Bicoid-interacting 3 (InterPro:IPR010675)
Overlap
AT5G54730 Medtr2g082770 AtATG18f (Arabidopsis thaliana homolog of yeast autophagy 18 (ATG18)
f)
Overlap
AT5G65280 Medtr3g086910 GCL1 (GCR2-LIKE 1); catalytic Overlap
ATMG00640 Medtr6g042930
Medtr6g043020
encodes a plant b subunit of mitochondrial ATP synthase based on
structural similarity and the presence in the F(0) complex.
Overlap
AT1G02060 Medtr7g044790 pentatricopeptide (PPR) repeat-containing protein TN113
AT1G04120 Medtr7g098690 ATMRP5 (Arabidopsis thaliana multidrug resistance-associated protein 5) TN113
AT1G06570 Medtr5g090980 PDS1 (PHYTOENE DESATURATION 1) TN113
AT1G07510 Medtr1g044600
Medtr4g064350
FTSH10 (FtsH protease 10); ATPase TN113
AT1G07530 Medtr2g097310
Medtr2g097350
Medtr2g097390
Medtr4g064160
scarecrow-like transcription factor 14 (SCL14) TN113
AT1G08460 Medtr2g087270 HDA08 (histone deacetylase 8); histone deacetylase TN113
AT1G08570 Medtr7g085310 thioredoxin family protein TN113
AT1G11020 Medtr4g113520 zinc finger (C3HC4-type RING finger) family protein TN113
AT1G11650 Medtr4g131970 ATRBP45B; RNA binding TN113
AT1G15740 Medtr3g043810
Medtr4g009870
Medtr4g116170
leucine-rich repeat family protein TN113
AT1G23070 Medtr2g042580 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT4G38360.2);
similar to unnamed protein product "Vitis vinifera" (GB:CAO65220.1);
contains InterPro domain Protein of unknown function DUF300
(InterPro:IPR005178)
TN113
AT1G26665 Medtr8g085570 similar to RNA polymerase II mediator complex protein-related TN113
AT1G27320 Medtr3g085130 AHK3 (ARABIDOPSIS HISTIDINE KINASE 3) TN113
AT1G32130 Medtr5g028030 similar to IWS1 C-terminus family protein TN113
AT1G35190 Medtr8g102600 oxidoreductase 2OG-Fe(II) oxygenase family protein TN113
129
AT1G35660 Medtr1g050670 binding TN113
AT1G36990 Medtr5g060280 similar to unnamed protein product "Vitis vinifera" TN113
AT1G47970 Medtr4g122830 unknown protein TN113
AT1G48280 Medtr4g124070 hydroxyproline-rich glycoprotein family protein TN113
AT1G50030 Medtr5g005380 TOR (TARGET OF RAPAMYCIN) TN113
AT1G53570 Medtr3g096230
Medtr8g093730
MAP3KA (Mitogen-activated protein kinase kinase kinase 3); kinase TN113
AT1G54710 Medtr1g082300 AtATG18h (Arabidopsis thaliana homolog of yeast autophagy 18 (ATG18)
h)
TN113
AT1G55325 Medtr3g083500 similar to hypothetical protein OsI_019475 "Oryza sativa (indica cultivar-
group)"
TN113
AT1G60200 Medtr5g034430 splicing factor PWI domain-containing protein / RNA recognition motif
(RRM)-containing protein
TN113
AT1G62290 Medtr4g132420 aspartyl protease family protein TN113
AT1G62710 Medtr4g101730 BETA-VPE (vacuolar processing enzyme beta); cysteine-type
endopeptidase
TN113
AT1G65950 Medtr5g077080 ABC1 family protein TN113
AT1G69220 Medtr5g045190 SIK1 (ERINE/THREONINE KINASE 1); kinase TN113
AT1G75440 Medtr5g021260
Medtr8g076490
UBC16 (UBIQUITIN-CONJUGATING ENZYME 16); ubiquitin-protein
ligase
TN113
AT1G76490 Medtr5g024880 HMG1 (3-HYDROXY-3-METHYLGLUTARYL COA REDUCTASE) TN113
AT1G78580 Medtr8g087910
Medtr8g087930
ATTPS1 (TREHALOSE-6-PHOSPHATE SYNTHASE); transferase
transferring glycosyl groups
TN113
AT2G02370 Medtr2g028450
Medtr3g064100
similar to unnamed protein product "Vitis vinifera" (GB:CAO22255.1);
contains InterPro domain SNARE associated Golgi protein
(InterPro:IPR015414)
TN113
AT2G17510 Medtr8g073170
Medtr8g073450
EMB2763 (EMBRYO DEFECTIVE 2763); RNA binding / ribonuclease TN113
AT2G32900 Medtr5g084830 ATZW10 TN113
AT2G35800 Medtr1g095780 mitochondrial substrate carrier family protein TN113
AT2G36350 Medtr3g010260
Medtr7g104180
protein kinase putative TN113
AT2G39190 Medtr1g095350
Medtr1g095480
ATATH8 (ABC2 homolog 8) TN113
AT2G39570 Medtr2g096640
Medtr7g024320
ACT domain-containing protein TN113
AT2G41710 Medtr7g091390 ovule development protein putative TN113
AT2G42520 Medtr3g083690 DEAD box RNA helicase putative TN113
AT2G44710 Medtr7g081210 RNA recognition motif (RRM)-containing protein TN113
AT2G45600 Medtr8g035520
Medtr8g035540
hydrolase TN113
AT2G45910 Medtr7g077780 protein kinase family protein / U-box domain-containing protein TN113
AT3G02290 Medtr7g016840 zinc finger (C3HC4-type RING finger) family protein TN113
AT3G02750 Medtr7g021530
Medtr7g080170
protein phosphatase 2C family protein / PP2C family protein TN113
AT3G02860 Medtr5g077800 zinc ion binding TN113
AT3G04590 Medtr1g079780
Medtr7g116320
DNA-binding family protein TN113
AT3G06480 Medtr2g032630 DEAD box RNA helicase putative TN113
AT3G06670 Medtr2g034560
Medtr4g121890
binding TN113
AT3G07890 Medtr5g098140 RabGAP/TBC domain-containing protein TN113
AT3G10690 Medtr1g031690 DNA gyrase subunit A family protein TN113
AT3G11760 Medtr1g056180 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT5G04860.1);
similar to unnamed protein product "Vitis vinifera" (GB:CAO39924.1)
TN113
130
AT3G15351 Medtr4g093680 similar to hypothetical protein OsI_025123 "Oryza sativa (indica cultivar-
group)" (GB:EAZ03891.1)
TN113
AT3G16830 Medtr4g120900 TPR2 (TOPLESS-RELATED 2) TN113
AT3G17970 Medtr8g107280 ATTOC64-III (ARABIDOPSIS THALIANA TRANSLOCON AT THE
OUTER MEMBRANE OF CHLOROPLASTS 64-III); binding / carbon-
nitrogen ligase with glutamine as amido-N-donor
TN113
AT3G19240 Medtr4g086240 similar to dem protein-related / defective embryo and meristems protein-
related
TN113
AT3G20720 Medtr2g082890
Medtr2g082910
Medtr4g081300
similar to hypothetical protein OsI_016901 "Oryza sativa (indica cultivar-
group)" (GB:EAY95668.1); similar to predicted protein "Physcomitrella
patens subsp. patens" (GB:EDQ56411.1)
TN113
AT3G20770 Medtr5g087790 EIN3 (ETHYLENE-INSENSITIVE3); transcription factor TN113
AT3G22490 Medtr1g072090
Medtr2g076230
late embryogenesis abundant protein putative / LEA protein putative TN113
AT3G26720 Medtr7g084040 glycosyl hydrolase family 38 protein TN113
AT3G51370 Medtr8g074930 protein phosphatase 2C putative / PP2C putative TN113
AT3G51620 Medtr5g075280 similar to nucleotidyltransferase family protein "Arabidopsis thaliana"
(TAIR:AT3G56320.1; contains domain TOPOISOMERASE-RELATED
PROTEIN (PTHR23092)
TN113
AT3G52850 Medtr7g073350
Medtr8g005800
ATELP/ATELP1/BP-80/BP80/BP80B/VSR-1/VSR1 (ARABIDOPSIS
THALIANA EPIDERMAL GROWTH FACTOR RECEPTOR-LIKE
PROTEIN)
TN113
AT3G54280 Medtr4g034920 ATP binding / DNA binding / helicase TN113
AT3G54850 Medtr1g093950 armadillo/beta-catenin repeat family protein / U-box domain-containing
family protein
TN113
AT3G58050 Medtr7g093630 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT2G41960.1) TN113
AT3G62770 Medtr1g083230
Medtr7g108520
AtATG18a (Arabidopsis thaliana homolog of yeast autophagy 18 (ATG18)
a)
TN113
AT4G02020 Medtr1g086980
Medtr7g109560
Medtr7g055680
EZA1 (SWINGER); transcription factor TN113
AT4G04670 Medtr4g132120 Met-10 like family protein / kelch repeat-containing protein TN113
AT4G05420 Medtr8g089380 DDB1A (UV-damaged DNA-binding protein 1A); DNA binding TN113
AT4G20850 Medtr1g101030 TPP2 (TRIPEPTIDYL PEPTIDASE II); subtilase TN113
AT4G21710 Medtr1g023500 NRPB2 (EMBRYO DEFECTIVE 1989); DNA binding TN113
AT4G26700 Medtr5g094160 ATFIM1 (Arabidopsis thaliana fimbrin 1); actin binding TN113
AT4G32160 Medtr3g106770 phox (PX) domain-containing protein TN113
AT4G32551 Medtr1g011610
Medtr4g113080
Medtr3g107910
Medtr1g082650
LUG (LEUNIG) TN113
AT4G32940 Medtr1g016590 GAMMA-VPE (Vacuolar processing enzyme gamma); cysteine-type
endopeptidase
TN113
AT4G35240 Medtr3g104120M
edtr5g098980
Medtr7g032230
similar to unknown protein "Arabidopsis thaliana" (TAIR:AT2G17110.1) TN113
AT4G35580 Medtr3g093040 no apical meristem (NAM) family protein TN113
AT4G35800 Medtr5g023020
Medtr5g077530
NRPB1 (RNA POLYMERASE II LARGE SUBUNIT); DNA binding /
DNA-directed RNA polymerase
TN113
AT4G36980 Medtr5g017430 similar to unnamed protein product "Vitis vinifera" (GB:CAO71849.1);
contains domain PTHR13161 (PTHR13161)
TN113
AT5G01990 Medtr7g074190 auxin efflux carrier family protein TN113
AT5G06210 Medtr3g025880 RNA-binding protein putative TN113
AT5G09880 Medtr3g095230
Medtr8g092550
RNA recognition motif (RRM)-containing protein TN113
AT5G10730 Medtr3g105660 binding / catalytic/ coenzyme binding TN113
AT5G10860 Medtr5g076010
Medtr5g076080
CBS domain-containing protein TN113
131
AT5G10940 Medtr3g100690 transducin family protein / WD-40 repeat family protein TN113
AT5G13300 Medtr7g020860 SFC (SCARFACE) TN113
AT5G16110 Medtr6g018430
Medtr6g084470
similar to unknown protein "Arabidopsis thaliana" (TAIR:AT3G02555.1);
similar to hypothetical protein "Cleome spinosa" (GB:ABD96917.1)
TN113
AT5G16210 Medtr7g082650 HEAT repeat-containing protein TN113
AT5G16880 Medtr4g077940
Medtr5g007480
Medtr4g078040
VHS domain-containing protein / GAT domain-containing protein TN113
AT5G19330 Medtr1g104870
Medtr5g005940
armadillo/beta-catenin repeat family protein / BTB/POZ domain-
containing protein
TN113
AT5G22220 Medtr4g052000 E2F1; transcription factor TN113
AT5G35200 Medtr4g115420 epsin N-terminal homology (ENTH) domain-containing protein TN113
AT5G35930 Medtr8g035620 AMP-dependent synthetase and ligase family protein TN113
AT5G39510 Medtr6g012860
Medtr7g011570
ATVTI11/ATVTI1A/SGR4/VTI11/VTI1A/ZIG (VESICLE
TRANSPORT V-SNARE 11); receptor
TN113
AT5G46840 Medtr5g029900 RNA recognition motif (RRM)-containing protein TN113
AT5G49710 Medtr5g065930 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT4G24590.1);
similar to unnamed protein product "Vitis vinifera" (GB:CAO61623.1)
TN113
AT5G51130 Medtr8g102950 contains InterPro domain Methyltransferase type 12 (InterPro:IPR013217);
contains InterPro domain Bicoid-interacting 3 (InterPro:IPR010675)
TN113
AT5G52660 Medtr7g038290 myb family transcription factor TN113
AT5G63860 Medtr3g096780
Medtr8g094410
Medtr8g094450
UVR8 (UVB-RESISTANCE 8) TN113
AT5G65530 Medtr5g031870 protein kinase putative TN113
ATMG00640 Medtr1g107010 encodes a plant b subunit of mitochondrial ATP synthase based on
structural similarity and the presence in the F(0) complex.
TN113
AT1G04300 Medtr1g044450 similar to meprin and TRAF homology domain-containing protein /
MATH domain-containing protein
TN115
AT1G08660 Medtr7g084010 glycosyl transferase family 29 protein / sialyltransferase family protein TN115
AT1G09780 Medtr7g074570 23-biphosphoglycerate-independent phosphoglycerate mutase putative /
phosphoglyceromutase putative
TN115
AT1G09960 Medtr5g067470 SUT4 (SUCROSE TRANSPORTER 4); carbohydrate transmembrane
transporter/ sucrose transmembrane transporter/ sucrose:hydrogen
symporter/ sugar:hydrogen ion symporter
TN115
AT1G10490 Medtr2g028030
Medtr8g092060
similar to predicted protein "Physcomitrella patens subsp. patens"; similar to
predicted protein "Nematostella vectensis"; contains InterPro domain
Protein of unknown function DUF699 ATPase putative
(InterPro:IPR007807); contains InterPro domain Region of unknown
function DUF1726 (InterPro:IPR013562)
TN115
AT1G24510 Medtr3g086330 T-complex protein 1 epsilon subunit putative / TCP-1-epsilon putative /
chaperonin putative
TN115
AT1G25380 Medtr5g055410 mitochondrial substrate carrier family protein TN115
AT1G31812 Medtr5g026740 ACBP (ACYL-COA-BINDING PROTEIN); acyl-CoA binding TN115
AT1G32360 Medtr8g066730 zinc finger (CCCH-type) family protein TN115
AT1G47128 Medtr1g018840 RD21 (RESPONSIVE TO DEHYDRATION 21); cysteine-type peptidase TN115
AT1G56110 Medtr4g103790 NOP56 (ARABIDOPSIS HOMOLOG OF NUCLEOLAR PROTEIN
NOP56)
TN115
AT1G65930 Medtr2g062840
Medtr5g077070
isocitrate dehydrogenase putative / NADP isocitrate dehydrogenase
putative
TN115
AT1G73230 Medtr3g020520
Medtr3g020660
Medtr4g071000
nascent polypeptide-associated complex (NAC) domain-containing protein TN115
AT1G78290 Medtr5g064540 serine/threonine protein kinase putative TN115
AT1G80530 Medtr2g104250
Medtr4g116210
nodulin family protein TN115
AT2G05990 Medtr4g074810
Medtr4g074950
MOD1 (MOSAIC DEATH 1); enoyl-"acyl-carrier-protein" reductase
(NADH)/ oxidoreductase
TN115
AT2G16460 Medtr5g021410 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT3G51090.1);
similar to Os06g0713100 "Oryza sativa (japonica cultivar-group)"
TN115
132
AT2G20450 Medtr1g075720
Medtr4g120770
60S ribosomal protein L14 (RPL14A) TN115
AT2G25110 Medtr3g106130
Medtr3g106160
Medtr1g012520
Medtr3g107040
Medtr3g107280
Medtr4g114650
SDF2 (STROMAL CELL-DERIVED FACTOR 2-LIKE PROTEIN
PRECURSOR)
TN115
AT2G25870 Medtr1g007320
Medtr1g007330
haloacid dehalogenase-like hydrolase family protein TN115
AT2G32060 Medtr4g006830
Medtr5g012890
Medtr8g093770
40S ribosomal protein S12 (RPS12C) TN115
AT2G34480 Medtr2g019720
Medtr2g098000
Medtr5g091120
60S ribosomal protein L18A (RPL18aB) TN115
AT2G38770 Medtr1g039070 EMB2765 (EMBRYO DEFECTIVE 2765) TN115
AT2G39890 Medtr3g069960
Medtr5g081630
ProT1 (PROLINE TRANSPORTER 1); amino acid transmembrane
transporter
TN115
AT2G40010 Medtr3g108280
Medtr5g082180
Medtr7g022400
60S acidic ribosomal protein P0 (RPP0A) TN115
AT2G41900 Medtr3g029590
Medtr7g092070
zinc finger (CCCH-type) family protein TN115
AT2G44060 Medtr5g088850 late embryogenesis abundant family protein / LEA family protein TN115
AT2G44120 Medtr1g108780 60S ribosomal protein L7 (RPL7C) TN115
AT2G44640 Medtr8g009290 similar to PDE320 (PIGMENT DEFECTIVE 320) "Arabidopsis thaliana"
(TAIR:AT3G06960.1); similar to unnamed protein product "Vitis vinifera"
(GB:CAO70456.1)
TN115
AT3G05060 Medtr5g010260 SAR DNA-binding protein putative TN115
AT3G05410 Medtr1g083110 similar to hypothetical protein OsI_004967 "Oryza sativa (indica cultivar-
group)" (GB:EAY77120.1); similar to hypothetical protein OsJ_004568
"Oryza sativa (japonica cultivar-group)" (GB:EAZ14743.1)
TN115
AT3G05560 Medtr1g083430
Medtr1g088450
60S ribosomal protein L22-2 (RPL22B) TN115
AT3G07760 Medtr5g097240 similar to unnamed protein product "Vitis vinifera" (GB:CAO41197.1);
contains InterPro domain Sterile alpha motif SAM (InterPro:IPR001660)
TN115
AT3G10690 Medtr1g031690 DNA gyrase subunit A family protein TN115
AT3G12110 Medtr2g008050
Medtr7g026230
ACT11 (ACTIN-11); structural constituent of cytoskeleton TN115
AT3G12930 Medtr2g069470 similar to unknown "Populus trichocarpa" (GB:ABK94112.1); contains
InterPro domain Iojap-related protein (InterPro:IPR004394)
TN115
AT3G18680 Medtr4g053800
Medtr8g093330
aspartate/glutamate/uridylate kinase family protein TN115
AT3G22300 Medtr1g005820
Medtr1g006120
RPS10 (RIBOSOMAL PROTEIN S10); structural constituent of ribosome TN115
AT3G48730 Medtr3g118070 GSA2 (GLUTAMATE-1-SEMIALDEHYDE 21-AMINOMUTASE 2);
glutamate-1-semialdehyde 21-aminomutase
TN115
AT3G52150 Medtr3g027140 RNA recognition motif (RRM)-containing protein TN115
AT3G53480 Medtr4g123850 ATPDR9/PDR9 (PLEIOTROPIC DRUG RESISTANCE 9); ATPase
coupled to transmembrane movement of substances
TN115
AT3G53740 Medtr1g100960 60S ribosomal protein L36 (RPL36B) TN115
AT3G54280 Medtr4g035100 ATP binding / DNA binding / helicase TN115
AT3G55280 Medtr2g096340
Medtr4g063060
60S ribosomal protein L23A (RPL23aB) TN115
AT3G56070 Medtr7g116340 ROC2 (rotamase CyP 2); peptidyl-prolyl cis-trans isomerase TN115
AT3G56150 Medtr1g023190
Medtr2g005610
Medtr7g116960
Medtr2g048870
Medtr6g045930
Medtr6g046050
EIF3C (EUKARYOTIC TRANSLATION INITIATION FACTOR 3) TN115
AT4G00100 Medtr4g102170
Medtr5g011910
ATRPS13A (RIBOSOMAL PROTEIN S13A); structural constituent of
ribosome
TN115
133
AT4G18480 Medtr2g015390 CHLI1 (CHLORINA 42); magnesium chelatase TN115
AT4G22380 Medtr4g114330 ribosomal protein L7Ae/L30e/S12e/Gadd45 family protein TN115
AT4G26210 Medtr5g062980 mitochondrial ATP synthase g subunit family protein TN115
AT4G26370 Medtr4g083370 antitermination NusB domain-containing protein TN115
AT4G27720 Medtr3g092030 similar to unknown protein "Arabidopsis thaliana" (TAIR:AT1G64650.2) TN115
AT4G27760 Medtr6g081020 FEY (FOREVER YOUNG); oxidoreductase TN115
AT4G29510 Medtr6g091770 ATPRMT11/PRMT11 (ARABIDOPSIS ARGININE
METHYLTRANSFERASE 11); protein-arginine N-methyltransferase
TN115
AT4G31340 Medtr3g109330 myosin heavy chain-related TN115
AT4G32605 Medtr5g093000 transcription factor TN115
AT4G35260 Medtr5g023740
Medtr8g074030
IDH1 (ISOCITRATE DEHYDROGENASE 1); isocitrate dehydrogenase
(NAD )
TN115
AT4G36420 Medtr8g076780 ribosomal protein L12 family protein TN115
AT4G37640 Medtr4g096990
Medtr5g015590
ACA2 (CALCIUM ATPASE 2); calmodulin binding TN115
AT5G01010 Medtr8g005040 similar to unnamed protein product "Vitis vinifera" (GB:CAO45344.1) TN115
AT5G04600 Medtr1g108290
Medtr4g127330
RNA recognition motif (RRM)-containing protein TN115
AT5G05010 Medtr2g101180 clathrin adaptor complexes medium subunit-related TN115
AT5G08160 Medtr4g112900 ATPK3 (Arabidopsis thaliana serine/threonine protein kinase 3); kinase TN115
AT5G08180 Medtr4g112780
Medtr5g077600
ribosomal protein L7Ae/L30e/S12e/Gadd45 family protein TN115
AT5G11420 Medtr1g011800
Medtr2g103170
similar to unknown "Ricinus communis" (GB:CAB02653.1); contains
InterPro domain Galactose-binding like (InterPro:IPR008979)
TN115
AT5G11950 Medtr1g015830 Encodes a protein of unknown function. It has been crystallized and shown
to be structurally almost identical to the protein encoded by At2G37210.
TN115
AT5G13300 Medtr8g100100 SFC (SCARFACE) TN115
AT5G13780 Medtr3g034200 GCN5-related N-acetyltransferase putative TN115
AT5G17020 Medtr4g077660
Medtr8g013700
Medtr5g007790
XPO1A (exportin 1A); protein transporter TN115
AT5G17710 Medtr7g087250 EMB1241 (EMBRYO DEFECTIVE 1241); adenyl-nucleotide exchange
factor/ chaperone binding / protein binding / protein homodimerization
TN115
AT5G18650 Medtr2g020040 zinc finger (C3HC4-type RING finger) family protein TN115
AT5G19750 Medtr7g088590 peroxisomal membrane 22 kDa family protein TN115
AT5G20180 Medtr4g114610 ribosomal protein L36 family protein TN115
AT5G20890 Medtr7g067460
Medtr7g113470
Medtr7g067470
chaperonin putative TN115
AT5G48630 Medtr7g055650 cyclin family protein TN115
AT5G50375 Medtr3g089010 CPI1 (CYCLOPROPYL ISOMERASE) TN115
AT5G51880 Medtr1g101840 oxidoreductase acting on paired donors with incorporation or reduction of
molecular oxygen 2-oxoglutarate as one donor and incorporation of one
atom each of oxygen into both donors
TN115
AT5G53070 Medtr5g038540 ribosomal protein L9 family protein TN115
AT5G62000 Medtr8g100050 ARF2 (AUXIN RESPONSE FACTOR 2); transcription factor TN115
AT5G64813 Medtr8g091980 LIP1 (LIGHT INSENSITIVE PERIOD1); GTPase TN115
AT5G66750 Medtr5g020000 CHR01/DDM1 (DECREASED DNA METHYLATION 1); helicase TN115
AT5G67570 Medtr4g097120 EMB1408 (EMBRYO DEFECTIVE 1408) TN115
134
Appendix B
The ecological genomic basis of salinity adaptation in
Tunisian Medicago truncatula
The following Research Article was accepted in BMC Genomics:
Friesen ML, von Wettberg EJB, Badri M, Moriuchi KS, Barhoumi F, Chang PL, Cuellar-
Ortiz S, Cordeiro MA, Vu WT, Arraouadi S, Djebali N, Zribi K, Badri Y, Porter SS, Aouani
ME, Cook DR, Strauss SY, Nuzhdin SV
Background: As our world becomes warmer, agriculture is increasingly impacted by rising soil
salinity and understanding plant adaptation to salt stress can help enable effective crop
breeding. Salt tolerance is a complex plant phenotype and we know little about the pathways
utilized by naturally tolerant plants. Legumes are important species in agricultural and natural
ecosystems, since they engage in symbiotic nitrogen-fixation, but are especially vulnerable to
salinity stress.
Results: Our studies of the model legume Medicago truncatula in field and greenhouse
settings demonstrate that Tunisian populations are locally adapted to saline soils at the
metapopulation level and that saline origin genotypes are less impacted by salt than
nonsaline origin genotypes; these populations thus likely contain adaptively diverged alleles.
Whole genome resequencing of 39 wild accessions reveals ongoing migration and candidate
genomic regions that assort non-randomly with soil salinity. Consistent with natural selection
acting at these sites, saline alleles are typically rare in the range-wide species' gene pool and
are also typically derived relative to the sister species M. littoralis. Candidate regions for
adaptation contain genes that regulate physiological acclimation to salt stress, such as abscisic
acid and jasmonic acid signaling, including a novel salt-tolerance candidate orthologous to
the uncharacterized gene AtCIPK21. Unexpectedly, these regions also contain biotic stress
genes and flowering time pathway genes. We show that flowering time is differentiated
between saline and non-saline populations and may allow salt stress escape.
Conclusions: This work nominates multiple potential pathways of adaptation to naturally
stressful environments in a model legume. These candidates point to the importance of both
tolerance and avoidance in natural legume populations. We have uncovered several
promising targets that could be used to breed for enhanced salt tolerance in crop legumes to
enhance food security in an era of increasing soil salinization.
135
Appendix C
Salinity adaptation and the contribution of parental effects in
Medicago truncatula
The following Research Article in currently in review in American Journal of Botany:
Ken S. Moriuchi,
Maren L. Friesen, Matilde A. Cordeiro,
Mounawer Badri, Wendy T. Vu,
Bradley J. Main, Mohamed Elarbi Aouani, Sergey V. Nuzhdin, Sharon Y. Strauss, and Eric
J.B. von Wettberg
Premise of the study: High soil salinity negatively influences plant growth and yield. Some taxa
have mechanisms for avoiding or tolerating elevated soil salinity, which can be modulated by
the environment experienced by parents or offspring. We examined potential mechanisms of
salinity adaptation and tested the contributions of parental and offspring environment on
salinity adaptation.
Methods: Salinity concentrations were factorially manipulated during parental and offspring
generations for the annual legume Medicago truncatula originating from populations from non-
saline and saline environments. We measured two aspects of plant performance,
reproduction and vegetative biomass, as well as, phenological and physiological traits
associated with salinity avoidance and tolerance mechanisms.
Key results: Saline-origin populations had greater biomass and reproduction under saline
conditions than non-saline populations, demonstrating adaptation of these populations to
saline soils. Parental environmental exposure to salt increased the difference in biomass of
saline and non-saline origin plants, and tended to do the same for reproduction, though this
difference was not significant. Salinity adaptation by saline origin populations was triggered
by environment. Parental exposure to salt spurred phenological differences that facilitated
salt avoidance. Offspring exposure to salt resulted in traits associated with greater water use
efficiency. Non-saline origin populations had traits associated with greater growth in the
absence of salt.
Conclusions: Plastic responses induced by parental and offspring environments in
phenological, leaf traits, and gas exchange have contributed to salinity adaptation in M.
truncatula. For saline adapted populations, the ability to maintain greater performance in
saline environments was also associated with lower growth potential in the absence of salt.
Abstract (if available)
Abstract
Genetic variation is essential for natural selection to operate and facilitate evolutionary processes. Identifying the mechanisms that influence the degree of genetic variation is important for our understanding of population differences and species diversity. Mechanisms that produce genetic variation are those that generate mutations in nucleotide sequences
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Complex mechanisms of cryptic genetic variation
PDF
Genetic architectures of phenotypic capacitance
PDF
Cellular level bottlenecks: genetic diversity, population dynamics, and technology development
PDF
Evolutionary genomic analysis in heterogeneous populations of non-model and model organisms
PDF
Physiological roles and evolutionary implications of alternative DNA polymerases in Escherichia coli
PDF
Application of genetic association methods in mice to understand phenotypes with a complex etiology
PDF
Understanding the genetics, evolutionary history, and biomechanics of the mammalian penis bone
PDF
Robustness and stochasticity in Drosophila development
PDF
Genetic and molecular insights into the genotype-phenotype relationship
PDF
Computational algorithms for studying human genetic variations -- structural variations and variable number tandem repeats
PDF
Behavioral choice assays and alcohol preference in Drosophila melanogaster
PDF
The molecular mechanism underlying the autoimmune-associated PTPN22 R620W variation and the quest for therapeutics
PDF
Response to alkylation damage linked to meiotic progression
PDF
Bayesian hierarchical models in genetic association studies
PDF
Natural variation of Arabidopsis thaliana methylome and its impact on genome evolution
PDF
Genome sequencing and transcriptome analysis of the phenotypically plastic spadefoot toads
PDF
From gamete to genome: evolutionary consequences of sexual conflict in house mice
PDF
Bayesian analysis of transcriptomic and genomic data
PDF
Innovative sequencing techniques elucidate gene regulatory evolution in Drosophila
PDF
The evolution of gene regulatory networks
Asset Metadata
Creator
Vu, Wendy T.
(author)
Core Title
Evolutionary mechanisms responsible for genetic and phenotypic variation
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Molecular Biology
Publication Date
08/11/2015
Defense Date
01/28/2015
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Drosophila,genetic variation,local adaptation,Medicago truncatula,OAI-PMH Harvest,salinity stress,seed,transcriptome,transgeneratational plasticity,transposable elements
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Nuzhdin, Sergey V. (
committee chair
), Ralph, Peter L. (
committee member
)
Creator Email
vu.wendy@gmail.com,wvu@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-632758
Unique identifier
UC11305764
Identifier
etd-VuWendyT-3824.pdf (filename),usctheses-c3-632758 (legacy record id)
Legacy Identifier
etd-VuWendyT-3824.pdf
Dmrecord
632758
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Vu, Wendy T.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
Drosophila
genetic variation
local adaptation
Medicago truncatula
salinity stress
seed
transcriptome
transgeneratational plasticity
transposable elements