Hawkmoths evaluate scenting flowers with the tip of their proboscis.

Alexander Haverkamp,Danny Kessler,Markus Knaden,Ian T Baldwin,Felipe Yon,Christine Mißbach,Ian W Keesey,Bill S Hansson,Christopher Koenig

doi:10.7554/elife.15039

Abstract

Pollination by insects is essential to many ecosystems. Previously, we have shown that floral scent is important to mediate pollen transfer between plants (Kessler et al., 2015). Yet, the mechanisms by which pollinators evaluate volatiles of single flowers remained unclear. Here, Nicotiana attenuata plants, in which floral volatiles have been genetically silenced and its hawkmoth pollinator, Manduca sexta, were used in semi-natural tent and wind-tunnel assays to explore the function of floral scent. We found that floral scent functions to increase the fitness of individual flowers not only by increasing detectability but also by enhancing the pollinator's foraging efforts. Combining proboscis choice tests with neurophysiological, anatomical and molecular analyses we show that this effect is governed by newly discovered olfactory neurons on the tip of the moth's proboscis. With the tip of their tongue, pollinators assess the advertisement of individual flowers, an ability essential for maintaining this important ecosystem service.

Highlights

Floral scent has been associated with insect pollination since the 18th century (Sprengler, 1793); the complex functions of floral volatiles have been only recently investigated in more detail, due to the availability of new molecular and analytical techniques (Raguso, 2008)
We investigated the function of floral scent in the context of individual flower-moth interactions, by offering individual male moths the choice between benzyl acetone (BA)-emitting flowers (i.e. empty-vector transformed flowers (EV)) and non-emitting flowers (i.e. CHAL) in a free flight tent (24 m Â 8 m Â 4 m, 10 CHAL and 10 empty vector control plants (EV) plants, spaced 50 cm apart)
The flight tracking revealed that moths chose to visit the same number of emitting and non-emitting flowers in a random sequence (Figure 1B, Figure 1— figure supplement 1A, probability of changing between EV or CHAL flowers during consecutive visits: 0.47)

Summary

Introduction

Floral scent has been associated with insect pollination since the 18th century (Sprengler, 1793); the complex functions of floral volatiles have been only recently investigated in more detail, due to the availability of new molecular and analytical techniques (Raguso, 2008). Research studying the function of floral scent has been divided along two themes with little cross-fertilization: 1) studies examining the fitness effects of floral scent without the causal behavioral responses of pollinators (Kessler et al, 2008), or 2) studies examining the sensory physiology of pollinators, neglecting the ecological consequences for the plant (Raguso, 2008). We meld these approaches and show that floral scent increases the fitness of individual flowers by increasing their detectability (Raguso and Willis, 2002), and by enhancing the pollinator’s foraging motivation, and demonstrate that this is mediated by olfactory receptors on the tip of the moth’s proboscis which detect floral scent. How floral scent emitted by individual flowers functions in this mutualistic interaction remained unknown, as it is unclear how pollinators detect single volatile compounds within complex natural environments (Hansson and Stensmyr, 2011; Riffell et al, 2014)

Results and discussion

Materials and methods

Funding Funder European Research Council