COMMENTARY COMMENTARY Functional loss of yeast detectors parallels transition to herbivory

Hany K. M. Dweck, Markus Knaden, and Bill S. Hansson1 paralleled by changes at the peripheral de- Department of Evolutionary Neuroethology, Max Planck Institute for Chemical Ecology, tection level or do the S. flava detect 07745 Jena, Germany the same set of odors like other drosophilids but exhibit a different behavioral pref- Herbivorous are extraordinarily suc- odors are produced by many yeast erence only due to shifts in higher-order cessful. Half of the world’sextantinsects and are strongly attractive to flies. Attraction olfactory processing? To address these and one quarter of all living metazoan species to the flies’ main food source (yeast) is thus questions, the authors test the antennal belong to herbivorous lineages (1, 2). to a large degree governed by these odorants responses of S. flava to the odors of its host Although feeding on living plant tissue is an and the involved receptors (7–10). However, plant Arabidopsis and of yeast and compare evolutionarily difficult transition, herbivo- some species of the genus Scaptomyza these responses with those of the yeast- rous insect lineages are more diverse than () have switched their food feeding D. melanogaster. In contrast to their nonherbivorous relatives (3, 4). In preference dramatically from microbes to D. melanogaster, antennae of S. flava respond PNAS, Goldman-Huertas et al. (5) present leaves and stems of living angiosperms (11). strongly to plant odors but give slight or no both the underlying genetic basis and the As this shift in preference has evolved only response to yeast volatiles. The peripheral resulting physiological changes that led to recently and as the now herbivorous flies odor detection machinery of S. flava there- ’ the evolution of herbivory in the fly Scap- are phylogenetically closely related to the fore reflects the flies trophic transition: the tomyza flava (Fig. 1). genomically most well-characterized and ge- flies did not only lose the preference for, but Most drosophilid flies including Drosoph- netically tractable species on earth, D. mel- also the ability to detect, most yeast volatiles, ila melanogaster feed on microbes growing ongaster, leaf mining Scaptomyza present whereas their sensitivity to some plant odors on decaying vegetation (6). This feeding pref- an excellent model to study the genomic was increased. Similar adaptive responses to erence is also mirrored in the flies’ olfactory and functional basis of the evolution of her- altered environmental conditions and shifts system, which is equipped with highly con- bivory in insects. in habitat preference have been observed in served odorant receptors (Ors) to detect yeast In well-designed experiments Goldman- the D. sechellia–Morinda (12) as well as in metabolites. Or9a and Or92a detect acetoin Huertas et al. show that the herbivorous fly the D. erecta–Pandanus associations (13). (7), Or42b detects ethyl acetate, and Or67a S. flava has lost the—in most other droso- It therefore seems to be a general rule that and Or85d detect phenyl ethanol and phenyl- philids highly conserved (6)—attraction to preference shifts and specialization come ethyl acetate, respectively (8, 9). All these yeast volatiles. Is this shift in preference along with shifts in the neural equipment of the flies’ olfactory sensory organs. The authors next look for chemosensory genes that could have been involved in the switch of preference and volatile detection. Antennal responses to odors are controlled by olfactory receptor (Or) genes expressed in olfactory sensory neurons housed in the sensilla on the flies’ antennae and palps. Goldman-Huertas et al. proceed to compare annotated Or genes in S. flava with those of closely related drosophilid species, in partic- ular those of D. melanogaster. It turns out that S. flava has lost a subset of the highly conserved Ors mentioned above (Or9a, Or42b, and Or85d), which are used by most drosophilid species to find fermenting host substrates. Interestingly, the ester-specific receptor, Or22a, which is tuned to Morinda- and Pandanus-characteristic volatiles in

Author contributions: H.K.M.D., M.K., and B.S.H. wrote the paper.

The authors declare no conflict of interest. See companion article on page 3026. Fig. 1. Emergence of herbivory in drosophilid flies based on the loss of yeast-specific receptors and resulting loss 1To whom correspondence should be addressed. Email: hansson@ attraction toward rotten fruits. ice.mpg.de.

www.pnas.org/cgi/doi/10.1073/pnas.1501319112 PNAS | March 10, 2015 | vol. 112 | no. 10 | 2927–2928 Downloaded by guest on September 25, 2021 D. sechellia and D. erecta,respectively(12,13), regarding the ecological significance of for the herbivorous fly S. flava. The se- has even been deleted in the herbivorous fly genes important in herbivory. Another ad- quenced genome of S. flava in combination S. flava. In addition, the authors find gene vantage of the interaction studied is the with the extensive genetic tools available duplication and hints for positive selection genetic tractability of the host plant, Arabi- for Arabidopsis could thus lift research on in one of the Ors of S. flava (Or67b). In dopsis, which is an important model for plant–insect interactions to a new level. The D. melanogaster, this receptor is responsible ecology and evolutionary biology (15). This, present study presents highly interesting for the detection of green leaf volatiles and in the future, will allow mutating the bio- results and forms an excellent vantage point is suggested to mediate aversion (14). Al- synthetic pathway of specific Arabidopsis for these different future directions of re- though the loss of attraction can be easily volatiles to characterize the ecological sig- search into the evolution of herbivory and explained by the loss of detection, the process nificance of different plant-produced signals of insect olfaction. of how formerly repellent odors can become attractive requires further investigations. 1 Ehrlich PR, Raven PH (1964) Butterflies and plants: A study in 9 Stökl J, et al. (2010) A deceptive pollination system targeting A further dissection of the S. flava olfac- coevolution. Evolution 18:586–603. drosophilids through olfactory mimicry of yeast. Curr Biol 20(20): tory system will also be highly interesting. 2 Bernays EA (1998) Evolution of feeding behavior in insect 1846–1852. herbivores. Bioscience 48:35–44. 10 Semmelhack JL, Wang JW (2009) Select glomeruli Have the numbers of detecting neurons 3 Mitter C, Farrell BD, Wiegmann B (1988) The phylogenetic study mediate innate olfactory attraction and aversion. Nature 459(7244): changed? Has the size relationship between of adaptive zones: Has phytophagy promoted insect diversification? 218–223. – olfactory glomeruli in the antennal lobe Am Nat 132:107 128. 11 Markow T, O’Grady P (2008) Reproductive ecology of 4 Farrell BD (1998) “Inordinate Fondness” explained: Why are there Drosophila. Funct Ecol 22(5):747–759. evolved and possibly now includes a macro- So many beetles? Science 281(5376):555–559. 12 Linz J, et al. (2013) Host plant-driven sensory specialization in 5 Goldman-Huertas B, et al. (2015) Evolution of herbivory in glomerular complex for the detection of Drosophila erecta. Proc Biol Sci 280(1760):20130626. Drosophilidae linked to loss of behaviors, antennal responses, green leaf odors? Also, maybe of highest 13 DekkerT,IbbaI,SijuKP,StensmyrMC,HanssonBS(2006) odorant receptors, and ancestral diet. Proc Natl Acad Sci USA Olfactory shifts parallel superspecialism for toxic fruit in interest, how have changes in wiring allowed 112:3026–3031. Drosophila melanogaster sibling, D. sechellia. Curr Biol 16(1): 6 Carson HL (1971) The ecology of Drosophila breeding sites. Lyon a circuit of previously negative valence to be- – Arboretum Lecture 2, ed Harold L (Univ of Hawaii, Honolulu), pp 1–28. 101 109. 14 come attractive. The study of Goldman- 7 Becher PG, Bengtsson M, Hansson BS, Witzgall P (2010) Flying the Stensmyr MC, Giordano E, Balloi A, Angioy AM, Hansson BS Huertas et al. provides new insights into fly: Long-range flight behavior of Drosophila melanogaster to (2003) Novel natural ligands for Drosophila olfactory receptor the role of loss of chemoreceptor genes in attractive odors. J Chem Ecol 36(6):599–607. neurones. J Exp Biol 206(Pt 4):715–724. 8 Dweck HKW, Ebrahim SAM, Farhan A, Hansson BS, Stensmyr MC 15 Mitchell-Olds T (2001) Arabidopsis thaliana and its wild relatives: the transition to herbivory. Furthermore, it (2015) Olfactory proxy-detection of dietary antioxidants in a model system for ecology and evoution. Trends Ecol Evol opens the door for further investigations Drosophila. Curr Biol 25(4):455–466. 16:693–700.

2928 | www.pnas.org/cgi/doi/10.1073/pnas.1501319112 Dweck et al. Downloaded by guest on September 25, 2021