Broadly speaking, our lab studies the genetic basis of phenotypic variation and evolution. Primarily using the fruit-fly, Drosophila melanogaster as a model system, we have explored questions with respect to trait variation (i.e. why do some individuals vary more than others for a given trait). In addition, we also exploring how traits co-vary together, and the genetic basis of this covariation. Specific areas of interest are described in more interest below.
Dissecting the genetic architecture of wing-shape in Drosophila. In evolutionary genetics, knowledge of the polymorphisms responsible for trait variation, and how they function (mean effect on a trait, interactions with each other and the environment, frequencies in populations), facilitates detailed analysis of trait evolution and tests of population genetic theory. To date, only a small number of candidate polymorphisms have been characterized, and beyond basic mapping, very little work has been done with respect to population genetics of such polymorphisms. In the lab we use wing-shape in Drosophila as a model trait to examine how evolutionary forces such as selection and drift shape the relationship between genotype and phenotype. In Drosophila, wing development has been the focus of research for over 60 years, and its genetic basis is reasonably well understood. In addition, there have numerous studies of intra and inter-specific variation for wing size and shape in Drosophila. While we understand how aspects of the wing develop, we do not understand how all of this information is integrated to make a wing of proper proportions in natural populations. During my recent work as a Post-doc with Greg Gibson, we initiated a number of studies to directly address these issue from two view points: First, with respect to known candidate genes that affect aspects of wing patterning and cell-fate determination, utilizing both naturally occurring polymorphisms and induced mutations that affect wing-shape. My lab is continuing this work, and as well we are utilizing wing-shape as a model system to study the evolution of DNA sequences that contribute to variation for transcription, and the relationship between variation in gene expression profiles (using micro-arrays), and variation for wing-shape.
1. Mapping naturally occurring polymorphisms
associated with trait variation.We have begun to
map naturally occurring polymorphisms that are associated with
variation for wing-shape. Previously I observed that a polymorphism in
a putative regulatory region of Epidermal growth
factor Receptor is associated with wing-shape in a
wild-caught cohort of flies (Dworkin et al. 2005). In the lab, we continue to use both experimental and "wild" flies to examine
functionally defined regulatory regions in known
“wing” genes to scan for naturally occurring
polymorphisms associated with variation for both wing-shape and gene
expression, as well as to study the molecular population genetics of
these enhancers.
2. Relationship between transcriptional and phenotypic variation To complement the numerous QTL mapping studies that have examined wing shape, We use induced mutations to further refine potential candidate genes that may harbour polymorphisms associated with natural variation for wing-shape. In a previous study, I selected mutations representing 40 genes in two signaling pathways involved with wing development. As a quantitative trait, wing-shape is sensitive to the effects of genetic background, therefore I introgressed the mutations into two common wild-type strains to study both the overall effects of the mutations, as well as the relative contribution of genetic background. Interestingly, the effects of the mutations are specific to different regions of the wing, as predicted by the effects of the patterning genes. We observed that mutations in genes from the same pathways do not necessarily have similar effects on wing-shape (Dworkin and Gibson 2006). Current work is focusing on the effect that these mutations have on transcriptional profiles, and how this in term is associated with variation for shape. Future work in this area will extend these studies to include epistatic relationships between genes in and among pathways.
Cryptic (Hidden) genetic
variation refers to the genetic component of total phenotypic variation
that is
revealed when a developmental programme is disrupted by an
environmental or
genetic perturbation. The most striking example of this kind of
variation was
demonstrated by C.H. Waddington (1956), who showed that when fly
embryos are
exposed to ether vapour, a small proportion show a homeotic
transformation (one
region or organ develops into another, like hands turning into feet).
When
Waddington selected upon individuals with this phenotype, in subsequent
generations, the homeotic phenotype was observed at higher frequencies,
suggesting that there was genetic variation for a phenotype, that had
likely
never been observed in nature. ![The effects of the sd[E3] in two different wild-type genetic backgrounds](images/Cryptic%20Genetic%20variation%20for%20Sd%20copy.jpg)
With respect to cryptic genetic variation, we are interested in why it is maintained (conditional neutrality?, pleiotropic effects of alleles?), and whether it plays a role in the maintenance of genetic variation for "observed" phenotypes in nature (See Gibson and Dworkin 2004). Previously, we mapped allelic variants associated with cryptic variation for the number of photoreceptors in the ommatidia of the fly eyes when genetically perturbed (Dworkin et al. 2003). In addition using a number of micro-array technologies, we are exploring the genomic consequences of background effects on gene expression. The figure above demonstrates how the same mutation (scallopedE3) can have profoundly different effects on the size of the wing depending upon the genetic background in which it is examined. Analysis of genome expression profiles suggest that there are a large number of genes whose transcription depends upon the interaction of genotype and background. In the future we hope to map the genetic modifiers that underly these background effects, to explore what (if any) role they may play in the maintainence of genetic variation in natural populations.
Canalization refers to a process of developmental homeostasis, as envisioned by C.H. Waddington (1942, 1952). The essential idea is that the developmental programme leading to trait expression is buffered against a degree of random genetic or environmental variation. This idea develops naturally from the observation that when such developmental systems break down via some form of perturbation, considerable phenotypic variation not observed under normal conditions is observed. More importantly, a portion of this variation is heritable suggesting cryptic genetic variation (see next section). One important question that has been addressed in a number of theoretical studies, is whether such developmental robustness evolves independently of the traits themselves.
My own work as a PhD student (under the guidance of Ellen Larsen) focused on exploring methods to infer canalization, and using Drosophila to test some theoretical models. I specifically set out to test a hypothesis developed by Gunter Wagner and colleagues (1997) that suggests that canalization of genetic effects has evolved as a correlated response to the evolution of environmental canalization. I studied this, and related questions by examining a number of traits including; sex-comb teeth and tarsus length under the influence of temperature and a mutation in Distal-less (Dworkin 2005a), as well as with sternopleural bristle number under the influence of the Sternopleural mutation (Dworkin 2005b).