We have an active research and breeding program that addresses the genetic and developmental basis of wing pattern evolution in Heliconius butterflies. In Heliconius, biologists are presented with a rare system in which the links between adaptation, morphological evolution, and development are well defined (TREE, 17:125-133). Our work has generally focused on H. erato, a species characterized by remarkable intraspecific diversification into over 20 parapatric color pattern races. The vivid wing patterns of H. erato are adaptations that warn potential predators of the butterflies' distastefulness. Recognizing the extraordinary potential of this species to be a model for understanding the links between development and evolution, NSF recently funded the development of a BAC library for H. erato (Nature 297, 1638-1639). Ongoing research on Heliconius is directed towards a number of specific genomic level questions including.

Wing pattern evolution in H. erato is a classic example of adaptation proceeding via changes at few genetic loci and serves as the foundation for current theories for the origin of mimicry. Major phenotypic changes in H. erato are clearly under the control a handful of major loci, so-called switch genes; however, the affects of these genes are complex and still poorly understood.

We have spent the last several years developing reference crosses and molecular markers for detailed analysis of the architecture of color pattern variation in H. erato. We now have large collections of F2 and backcross broods for phenotypic and genetic analysis. We have also developed an extensive suite of co-dominant loci for our mapping work. Co-dominant loci are essential anchors for integrating across AFLP maps and for comparative genomic work.

The first-generation linkage map is now complete (Molecular Ecology, in prep) and we are in the process of constructing the second and third-generation maps. Our overriding goal is to generate high-resolution chromosomal linkage groups containing all major patterning loci and to use this information to i) better understand color pattern evolution within H. erato, ii) map the position of potential candidate loci relative to major color pattern switch genes, iii) determine the conservation of color pattern linkage groups across Heliconius and, iv) ultimately, to identify the pattern loci themselves (now realistic with access to BAC library resources for H. erato). To that end we have automated AFLP analysis and scored more than 700 AFLP loci in our reference F2 brood. In addition, we are developing methods for rapidly scoring single nucleotide polymorphisms in co-dominant markers. Furthermore, we have are beginning collaborative research with Dr. Chris Jiggins (STRI/University of Edinburgh). One of the principle aims of this collaboration is to examine the conservation of linkage blocks in H. erato and H. melpomene. Heliconius erato and H. melpomene are a particularly fascinating species pair to begin comparative genomic work. These species represent two of the most divergent lineages within Heliconius, yet they have nearly identical wing pattern phenotypes and display parallel geographic patterns of divergence in their aposematic wing color patterns. The striking similarities in the color patterns of the distantly related H. erato and H. melpomene have lead some researchers to speculate that homologous loci must be involved in pattern generation in the two species. By contrasting the linkage relationships between the color pattern genes and homologous genetic markers, we hope to provide genetic evidence either in support of or against this intriguing hypothesis.

In collaboration with Dr. Fred Nijhout at Duke, we are beginning functional genomic analysis of wing pattern development in H. erato. The goal of this research is to extend the history of ecological and evolutionary research on Heliconius and utilize advances in gene chip technology to place Heliconius butterflies squarely at the interface between developmental genetics and evolution. In many respects, the color patterns of Lepidoptera wings are ideal model systems to apply new empirical and analytical methods to understand how changes in gene expression generate phenotypic variation. Wing patterns are structurally simple, highly variable, and often under strong and identifiable selection. Importantly, the development of colored scale cells and the spatial coordinate system for patterning the wing are also relatively recent evolutionary novelties. As a result, the genetic networks that control pattern development and diversification are likely to be simple relative to more complex morphological traits.

We are in the very early stages of this collaborative effort. We have developed a good wing disc library for H. erato and identified 500+ unique ESTs from an initial screen of this library. We are now making a ³mini-chip² composed of 100 developmental and pigment loci and using it to study gene expression. Ultimately, we will develop a Heliconius chip composed of all the genes expressed during wing development and use this powerful new resource to determine the time series of gene expression across differently colored areas of the wing and in different color pattern races of H. erato. These data will allow us to reconstruct the gene networks that mediate pigment production and understand how these networks are modified during development and morphological diversification.