Electrophysiological and behavioural characterization of gustatory responses to antennal ‘bitter’ taste in honeybees Maria Gabriela de Brito Sanchez, Martin Giurfa, Theo Rolla de Paula Mota and Monique Gauthier Centre de Recherches sur la Cognition Animale (UMR 5169), CNRS – Universite ´ Paul Sabatier, 118 route de Narbonne, 31062 Toulouse cedex 4, France Keywords: antenna, Apis mellifera, bitter taste, gustation, honeybee Abstract We combined behavioural and electrophysiological experiments to study whether bitter taste is perceived at the antennal level in honeybees, Apis mellifera. Our behavioural studies showed that neither quinine nor salicin delivered at one antenna at different concentrations induced a retraction of the proboscis once it was extended in response to 1 m sucrose solution delivered to the opposite antenna. Bees that extended massively their proboscis to 1 m sucrose responded only partially when stimulated with a mixture of 1 m sucrose and 100 mm quinine. The mixture of 1 m sucrose and 100 mm salicin had no such suppressive effect. No behavioural suppression was found for mixtures of salt solution and either bitter substance. Electrophysiological recordings of taste sensillae at the antennal tip revealed sensillae that responded specifically either to sucrose or salt solutions, but none responded to the bitter substances quinine and salicin at the different concentrations tested. The electrophysiological responses of sensillae to 15 mm sucrose solution were inhibited by a mixture of 15 mm sucrose and 0.1 mm quinine, but not by a mixture of 15 mm sucrose and 0.1 mm salicin. The responses of sensillae to 50 mm NaCl were reduced by a mixture of 50 mm NaCl and 1 mm quinine but not by a mixture of 50 mm NaCl and 1 mm salicin. We concluded that no receptor cells for the bitter substances tested, exist at the level of the antennal tip of the honeybee and that antennal bitter taste is not represented as a separate perceptual quality. Introduction Bitter substances are biologically relevant due to their potential toxicity. It is therefore not surprising that mammals have specialized cells to detect them (Scott, 2004; Mueller et al., 2005). Recent molecular studies suggest that receptor molecules for sweet and bitter taste are expressed in different cells (Hoon et al., 1999; Adler et al., 2000; Nelson et al., 2001; Zhang et al., 2003). In insects, taste transduction is performed by gustatory receptor neurons, which are held within cuticular hairs called taste sensillae, which have a pore on the tip. Taste sensillae are located essentially on the mouth parts, the legs, the wing margins, and some parts of the thorax and the abdomen (reviewed in Chapman, 2003). Gustatory receptor cells specifically tuned to ‘bitter’ substances have been found recently in the fruit fly Drosophila melanogaster . [Although the use of the term ‘bitter’ constitutes an anthropocentrism in the case of insects, we use it here as is usually done in the insect gustatory literature (Tanimura & Kikuchi, 1972; Glendinning et al., 2001, 2002)]. Meunier et al. (2003) showed that some taste sensilla on the terminal tarsal segments of the prothoracic legs detect bitter substances both through the activation of taste neurons but also through the inhibition of taste neurons activated by sugars and water. A specific receptor neuron within such sensilla, the L2 cell, was suggested as a candidate for responding selectively to bitter substances and to salt at high concentrations (Meunier et al., 2003). Hiroi et al. (2004) also reported sensitivity to bitter substances in sensilla at the periphery of the labellum of Drosophila, i.e. close to the proboscis. The receptor neuron responding to bitter taste also responds to salt at high concentrations (Hiroi et al., 2004). Bitter taste sensitivity seems to be confined in the case of Drosophila to the tarsal segments of the prothoracic legs and to the distal region of the proboscis (Wang et al., 2004). This makes sense as gustatory exploration occurs via the legs, which contact the substrate when the fly stands or walks, and via the labellum, which contacts the substrate when the insect feeds. Other organs such as the antennae are extremely short in the case of Drosophila and may not serve food sensing as efficiently as the legs and the labellum. In the honeybee Apis mellifera, the antennae are long and, together with the mouthparts and the fore tarsi, constitute the main chemosen- sory organs (Mitchell et al., 1999; Goodman, 2003). Gustatory antennal reception plays a role in appetitive food sensing as shown by the fact that stimulation of the antennae with sucrose solution elicits the so-called proboscis extension reflex (PER; Takeda, 1961; Bitterman et al., 1983). Less is known, however, about how gustatory information is processed at the periphery of the honeybee nervous system. Previous electrophysiological studies have focused essentially on sucrose specialized sensillae located on the mouth parts (Whitehead & Larsen, 1976; Whitehead, 1978) and the antennae (Haupt, 2004). However, no study has analysed the issue of peripheral coding of bitter taste in the honeybee. Contradictory behavioural evidence exists with respect to the sensitivity of bees to bitter substances. von Frisch (1967) reported that honeybees are insensitive to bitter substances. However, Chittka et al. (2003) reported that bumble bees are highly sensitive to quinine stimulation, which seems to be aversive and to improve the learning of visual discriminations in which one colour is rewarded Correspondence: Dr M. G. de Brito Sanchez, as above. E-mail: debrito@cict.fr Received 13 June 2005, revised 24 October 2005, accepted 30 October 2005 European Journal of Neuroscience, Vol. 22, pp. 3161–3170, 2005 ª Federation of European Neuroscience Societies doi:10.1111/j.1460-9568.2005.04516.x