Local adaptation of gene regulation in natural populations of Drosophila melanogaster

The central goal of this dissertation is to understand the genetic
and functional aspects of how populations adapt to new or changing
environments. Genetic variation within a population, either at
protein coding genes or at regulatory elements, provides the
substrate upon which natural selection can act to drive adaptation.
There is considerable evidence that changes in gene expression
account for a large proportion of morphological, physiological and
behavioral variation between and within species that can contribute
to adaptation and speciation. Due to the major role that gene
expression changes can have in shaping phenotypes, the first three
chapters of this dissertation deal with the study of how changes in
gene expression can facilitate adaptation. I use Drosophila
melanogaster from ancestral and derived regions of the species'
range as a model system for studying local adaptation. In chapter
1, I perform high-throughput RNA-sequencing (RNA-seq) of brain
tissue of flies from an ancestral (Zimbabwe) and a derived (the
Netherlands) population. The whole brain transcriptome was assayed
for differences in gene expression between African and European
flies in order to understand how differences in brain expression
may lead to local adaptation. I found over 300 candidate genes that
differed significantly in expression between the populations,
including a cluster of genes on chromosome arm 3R that showed
reduced expression in Europe and genetic evidence for positive
selection. Other candidate genes involved in stress response,
olfaction and detoxification were also identified. Additionally, I
compared brain gene expression between males and females and found
an enrichment of sex-biased genes on the X chromosome. Chapter 2
presents a detailed study of one of the candidate genes identified
in chapter 1. The metallothionein gene, MtnA, shows over four-fold
higher expression in the brain of European flies than of African
flies. I found a derived deletion in the 3’ untranslated region
(UTR) of MtnA that segregates at high frequency within the Dutch
population, but is absent from the Zimbabwean population. The
presence of the deletion was perfectly associated with the observed
variation in MtnA expression. When additional populations of D.
melanogaster were screened for the presence of the deletion, I
found that it showed a clinal distribution that was significantly
correlated with latitude and temperature. Furthermore, using
population genetic data and a selective sweep analysis I show that
the MtnA locus is evolving under positive selection. In Chapter 3 I
report a population genetic analysis of the enhancer region of
CG9505, a gene that shows significantly higher expression in
European than in African populations of D. melanogaster. A previous
study found that there was very low nucleotide polymorphism in the
enhancer region of CG9509 in flies from the Netherlands, a pattern
that is consistent with a selective sweep. I analyzed an additional
set of five populations from Zambia, Egypt, Malaysia, France and
Germany in order to gain a better understanding of how selection
has affected the evolution of this enhancer. I found that there is
a depletion of nucleotide diversity in all of the non-sub-Saharan
populations (Egypt, Malaysia, France and Germany), which share a
common high-frequency derived haplotype. Population genetic
analyses suggest that a selective sweep took place in the enhancer
region of CG9509 just after D. melanogaster migrated out of
sub-Saharan Africa. Finally, in chapter 4 I performed in situ
hybridizations to examine the expression of tissue-specific
reporter genes in the D. melanogaster testis. In the male germline
of D. melanogaster, reporter genes that reside on the X chromosome
show a reduction in expression relative to those located on the
autosomes. This phenomenon was demonstrated by randomly inserting
reporter gene constructs on the X chromosome and the autosomes. By
doing in situ hybridizations on testis of flies having reporter
gene insertions on the X chromosome and autosomes, I could show
that the expression difference mainly occurs in meiotic and
post-meiotic cells. For most constructs, expression was very low or
absent in the testis apex, which is enriched with pre-meiotic
cells. These results suggest that the suppression of X-linked gene
expression in the Drosophila male germline occurs through a
different mechanism than the MSCI (meiotic sex chromosome
inactivation) known to occur in mammals.