Rapid Microsatellite Isolation from a Butterfly by De Novo Transcriptome Sequencing: Performance and a Comparison with AFLP-Derived Distances Alexander S. Mikheyev 1 *, Tanya Vo 1 , Brian Wee 2 , Michael C. Singer 3 , Camille Parmesan 3 1 Okinawa Institute of Science and Technology, Onna, Japan, 2 National Ecological Observatory Network, Incorporated, Washington, District of Columbia, United States of America, 3 University of Texas, Section of Integrative Biology, Austin, Texas, United States of America Abstract Background: The isolation of microsatellite markers remains laborious and expensive. For some taxa, such as Lepidoptera, development of microsatellite markers has been particularly difficult, as many markers appear to be located in repetitive DNA and have nearly identical flanking regions. We attempted to circumvent this problem by bioinformatic mining of microsatellite sequences from a de novo-sequenced transcriptome of a butterfly (Euphydryas editha). Principal Findings: By searching the assembled sequence data for perfect microsatellite repeats we found 10 polymorphic loci. Although, like many expressed sequence tag-derived microsatellites, our markers show strong deviations from Hardy- Weinberg equilibrium in many populations, and, in some cases, a high incidence of null alleles, we show that they nonetheless provide measures of population differentiation consistent with those obtained by amplified fragment length polymorphism analysis. Estimates of pairwise population differentiation between 23 populations were concordant between microsatellite-derived data and AFLP analysis of the same samples (r = 0.71, p,0.00001, 425 individuals from 23 populations). Significance: De novo transcriptional sequencing appears to be a rapid and cost-effective tool for developing microsatellite markers for difficult genomes. Citation: Mikheyev AS, Vo T, Wee B, Singer MC, Parmesan C (2010) Rapid Microsatellite Isolation from a Butterfly by De Novo Transcriptome Sequencing: Performance and a Comparison with AFLP-Derived Distances. PLoS ONE 5(6): e11212. doi:10.1371/journal.pone.0011212 Editor: Bengt Hansson, Department of Animal Ecology, Lund University, Sweden Received March 15, 2010; Accepted April 26, 2010; Published June 18, 2010 Copyright: ß 2010 Mikheyev et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work has been funded by a grant from the United States Fish and Wildlife Service. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: alexander.mikheyev@oist.jp Introduction Many types of genetic analysis take advantage of microsatellite markers, which are highly polymorphic loci of simple sequence repeats located through the genome. For example, microsatellite analysis is useful in studies of paternity, population structure and history, as well as to make conservation decisions for the management of endangered species [1,2]. Given the broad-scale utility of these markers, a large number of approaches have been developed for their isolation from genomic DNA [3]. These approaches typically involve some form of microsatellite enrichment, followed by time consuming and costly brute force sequencing. Aside for the labor and cost associated with traditional approaches, the microsatellite enrichment step sometimes fails. For example, for reasons not fully understood, isolation of microsatellites from Lepidopteran genomes is ex- tremely difficult [4–6]. This problem is not confined to Lepidoptera, affecting bivalve mollusks [7], mosquitoes [8], mites [9], ticks [8], nematodes [10,11] and birds [12,13]. The increase in publicly available EST data for many species has made bioinformatic isolation of microsatellite markers increasingly commonplace (e.g., [14–17]). However, microsatellites isolated from EST libraries differ from those typically found in regions of the genome unassociated with genes. Gene-associated microsatellites are physically linked to particular alleles of a gene, and may hitchhike if the gene is under selection. Microsatellite variation in untranslated regions of transcribed DNA may affect the rates of gene expression or translation, and thus may be under selection. Indeed, EST-derived microsatellites almost universally show strong deviations from Hardy-Weinberg equilibrium. However, the relatively few studies that compare the performance of EST-derived microsatellites with that of other genotyping techniques have generally found comparable results [18–22]. Here we used the Roche 454 Titanium platform for transcriptional sequencing of Edith’s checkerspot butterfly (Euphydryas editha), in order to rapidly isolate polymorphic microsatellite loci for a conservation genetics study. We then compared the estimates of population differentiation and biogeographic structure obtained by this approach with those from AFLP genotyping of the same set of populations [23]. Materials and Methods Microsatellite identification RNA was extracted from a larva, a pupa and an adult E. editha. RNA extraction, normalized library preparation, sequencing and PLoS ONE | www.plosone.org 1 June 2010 | Volume 5 | Issue 6 | e11212