Genes Involved in the Evolution of Herbivory by a Leaf-Mining, Drosophilid Fly
Total Page:16
File Type:pdf, Size:1020Kb
GBE Genes Involved in the Evolution of Herbivory by a Leaf-Mining, Drosophilid Fly Noah K. Whiteman1,2,*, Andrew D. Gloss1,y, Timothy B. Sackton2,y, Simon C. Groen2,8, Parris T. Humphrey1, Richard T. Lapoint1, Ida E. Sønderby3,9, Barbara A. Halkier3, Christine Kocks4,5,6,10, Frederick M. Ausubel5,7,y, and Naomi E. Pierce2,y 1Department of Ecology and Evolutionary Biology, University of Arizona 2Department of Organismic and Evolutionary Biology and Museum of Comparative Zoology, Harvard University 3Department of Plant Biology and Biotechnology, Faculty of Life Sciences, University of Copenhagen, Denmark 4Department of Pediatrics, Massachusetts General Hospital, Boston, Massachusetts 5Department of Molecular Biology, Massachusetts General Hospital, Boston, Massachusetts 6Department of Pediatrics, Harvard Medical School 7Department of Genetics, Harvard Medical School 8Present address: Department of Plant Sciences, University of Cambridge, United Kingdom 9Present address: Department of Medical Genetics, Oslo University Hospital, Ulleva˚l and Institute of Psychiatry, University of Oslo, Norway 10Present address: Max Delbrueck Center for Molecular Medicine (MDC), Berlin, Germany *Corresponding author: E-mail: [email protected]. yThese authors contributed equally to this work. Accepted: July 16, 2012 Data deposition: All Sanger-sequences have been deposited in GenBank under the accession number JX160018-JX160047, and all short-read sequences have been deposited through the Short Read Archive at NCBI under the accession number SRA054250. Abstract Herbivorous insects are among the most successful radiations of life. However, we know little about the processes underpinning the evolution of herbivory. We examined the evolution of herbivory in the fly, Scaptomyza flava, whose larvae are leaf miners on species of Brassicaceae, including the widely studied reference plant, Arabidopsis thaliana (Arabidopsis). Scaptomyza flava is phylogenetically nested within the paraphyletic genus Drosophila, and the whole genome sequences available for 12 species of Drosophila facilitated phylogenetic analysis and assembly of a transcriptome for S. flava. A time-calibrated phylogeny indicated that leaf mining in Scaptomyza evolved between 6 and 16 million years ago. Feeding assays showed that biosynthesis of glucosinolates, the major class of antiherbivore chemical defense compounds in mustard leaves, was upregulated by S. flava larval feeding. The presence of glucosinolates in wild-type (WT) Arabidopsis plants reduced S. flava larval weight gain and increased egg–adult development time relative to flies reared in glucosinolate knockout (GKO) plants. An analysis of gene expression differences in 5-day-old larvae reared on WT versus GKO plants showed a total of 341 transcripts that were differentially regulated by glucosinolate uptake in larval S. flava. Of these, approximately a third corresponded to homologs of Drosophila melanogaster genes associated with starvation, dietary toxin-, heat-, oxidation-, and aging-related stress. The upregulated transcripts exhibited elevated rates of protein evolution compared with unregulated transcripts. The remaining differentially regulated transcripts also contained a higher proportion of novel genes than the unregulated transcripts. Thus, the transition to herbivory in Scaptomyza appears to be coupled with the evolution of novel genes and the co-option of conserved stress-related genes. Key words: Arabidopsis, Drosophila, glucosinolates, herbivory, host specialization, transcriptome. Introduction living insect species and 25% of living metazoan species Phytophagous insects are among the most evolutionarily suc- (EhrlichandRaven1964; Bernays 1998)andaremorediverse cessful radiations. Herbivorous taxa comprise almost 50% of than their aphytophagous sister lineages (Mitter and Farrell ß The Author(s) 2012. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. 900 Genome Biol. Evol. 4(9):900–916. doi:10.1093/gbe/evs063 Advance Access publication July 19, 2012 Genes Involved in the Evolution of Herbivory GBE 1988; Farrell 1998). Adapting to a herbivorous lifestyle is 2005; Dworkin and Jones 2009), it seems unlikely that host thought to have been a difficult transition for most insects, plants or fungi would mount a defense response against these however, due to morphological and physiological challenges insects. However, drosophilids in the genus Scaptomyza of consuming living plant tissues that are typically well de- (fig. 1) have evolved the ability to feed on living angiosperm fended against insects (Frankel 1959; Ehrlich and Raven leaves, which activate inducible defenses in response to feed- 1964; Southwood 1973). Some antiherbivore defenses, in- ing damage (Hackman 1959; Whiteman et al. 2011). It is likely cluding insecticidal compounds produced by plants, are dra- that a variety of preformed and inducible defense compounds matically induced within hours of herbivore attack and are were novel traits encountered by Scaptomyza and other her- tailored toward specific guilds of insect enemies (Kunitz bivorous Drosophilidae in their transition from a saprophagous 1945; Bostock 2005; Textor and Gershenzon 2009). In addi- to a phytophagous lifestyle. tion, these compounds are often effective against other nat- The ancestral host plants of the earliest Scaptomyza herbi- ural enemies, such as microbes (Fan et al. 2011), and vores are unknown, but many herbivorous Scaptomyza spe- vertebrates (Magnanou et al. 2009). cies are monophagous or oligophagous on plants in the order The genomic changes that accompany a shift from an Brassicales (Hackman 1959). The larvae of S. flava feed on aphytophagous to phytophagous lifestyle are not well under- Arabidopsis thaliana (hereafter referred to as Arabidopsis) stood. The transition to herbivory has evolved independently and other species of Brassicales in the wild (Chittenden many times in the insects and frequently involved adaptation 1902; Whiteman et al. 2011) and require living plants to com- to a more specialized host plant range. Indeed, most herbiv- plete their life cycle (Whiteman et al. 2011). Because they have orous species are specialized in host range and are only able to been found on many members of the Brassicales and on spe- complete development on a limited number of species. Many cies in a few other plant families (Caryophyllaceae and evolutionary changes are likely to be required for a successful Fabaceae), we consider this species to be oligophagous transition to herbivory, including the ability to find suitable (Martin 2004). We took advantage of the availability of host plants, utilize potentially nutritionally imbalanced plant Arabidopsis mutants blocking the synthesis of insecticidal tissues, and cope with host plant defenses (Bernays and compounds to test the hypothesis that major changes associ- Chapman 1994; Govind et al. 2010). ated with the transition to feeding on living plants can be To study the transition to herbivory from an evolutionary detected at the genomic level in S. flava. perspective, we have focused on a recently derived herbivo- We focused our attention on interactions between larval rous species that is closely related to a model genetic species. S. flava and glucosinolates, which are well characterized func- Our long-term goal is to understand how the evolution of tionally and metabolically in Arabidopsis (Halkier and novel genes and/or the recruitment of existing metabolic or Gershenzon 2006) and which we hypothesized were a likely signaling pathways have enabled herbivores to adapt to a barrier to colonization by ancestors of S. flava. Glucosinolates fundamentally new niche. Most phytophagous insects are lep- are a major barrier to insect colonization of Brassicales, includ- idopterans or coleopterans. Because the evolution of herbivory ing economically important crops such as canola (Brassica in these lineages is estimated to have occurred in the napus), broccoli, cabbage (B. oleracea), and papaya (Carica Cretaceous, 145–65 million years ago (Whalley 1977; papaya). Glucosinolates (and related compounds called cama- Moran 1989; Labandeira et al. 1994; Bernays 1998; Farrell lexins [Hull et al. 2000]) are amino acid-derived thioglucosides 1998), using comparative genomics to dissect the genetic that are present constitutively in tissues of plants in the bases of these transitions would be difficult. However, herbiv- Brassicaceae and are highly inducible after herbivore attack, ory has also evolved many times in dipterans (Labandeira increasing in concentration in leaves by up to 40-fold (Halkier 2003), and the family Drosophilidae includes 12 species with and Gershenzon 2006; Textor and Gershenzon 2009). During published genome sequences and at least two genera tissue damage, glucosinolates are hydrolyzed by endogenous (Scaptomyza and Scaptodrosophila) with phytophagous myrosinases and are transformed into toxic, electrophilic mus- members (Hackman 1959; Bock and Parsons 1975; Clark tard oils (isothiocyanates) that can damage proteins and DNA et al. 2007). These provide the opportunity to analyze more (Halkier and Gershenzon 2006). recent transitions to herbivory. We used a variety of approaches,