Bees rely on amino acids from nectar and pollen for essential physiological functions. While nectar typically contains low (<1 mM) amino acid concentrations, levels in pollen are higher but variable (10-200 mM). Behavioural studies suggest bumblebees have preferences for specific amino acids but whether such preferences are mediated via gustatory mechanisms remains unclear. This study explores bumblebees' (Bombus terrestris) gustatory sensitivity to two essential amino acids found in nectar and pollen, valine and lysine, using electrophysiological recordings from gustatory sensilla on their mouthparts. Valine elicited a concentration-dependent response from 0.1 mM, indicating that bumblebees could perceive valine at concentrations found naturally in nectar and pollen. In contrast, lysine failed to evoke a response across tested concentrations (0.1-500 mM). The absence of lysine detection raises questions about the specificity and diversity of amino acid-sensitive receptors in bumblebees. Bees responded to valine at lower concentrations than sucrose, suggesting comparatively higher sensitivity (EC₅₀: 0.7 mM vs. 3.91 mM for sucrose). Our findings indicate that bumblebees can evaluate the amino acid content of pollen and nectar using pre-ingestive cues, rather than relying on post-ingestive cues or feedback from their nestmates. Such sensory capabilities likely impact foraging strategies, with implications for plant-bee interactions and pollination.

Bumblebee (Bombus terrestris audax) colonies were obtained from Biobest (Westerlo, Belgium) (n=2) and Koppert (Berkel en Rodenrijs, The Netherlands) (n=1). Colonies were maintained at the University of Sussex, U.K., housed either within a flight cage (73x73x65 cm) or connected to a feeding arena (40x40x35 cm) via a corridor (4x4x26 cm). Bees had ad libitum access to a nectar substitute (Biogluc, Biobest, Westerlo, Belgium) via both ground and suspended feeders, and finely ground honeybee collected pollen (Agralan, Swindon, U.K.), presented on chenille stems placed inside white plastic cups. Nectar was replenished as needed and pollen changed daily. Workers observed collecting pollen from the feeders were marked on the thorax with a small dot of white enamel paint (Humbrol, Hornby Hobbies Limited, Margate, U.K.). Only individuals observed collecting pollen at least once were used in testing (n=27).

Bumblebees (n=27) were caught in the feeding arena and left at 4 °C overnight to immobilise them. On the day of recording, bees were inserted in small plastic tubes and harnessed using small strips of Parafilm M (American National Can, Greenwich, CT, U.S.A.). Bees were then transferred to a plate of sealing wax beneath a microscope (Nikon AZ100, Tokyo, Japan). Using wax, the head, antennae, and front legs of the bumblebee were immobilised, and the glossa and labial palps were manually extended and immobilised to the plastic tube. The galeae were rinsed in ultrapure water and dried with QL100 filter paper (Fisherbrand, Fisher Scientific, Loughborough, U.K.), and subsequently extended and fixed to the plastic tube using small strips of Parafilm to prevent heat damage.

Extracellular tip recordings (figure 1a) were performed on A-type sensilla chaetica on the left galea [27]. A 25 μm tungsten wire (Alfa Aesar, Ward Hill, MA, U.S.A.) was inserted into the galea at the proximal end and pushed gently down to ~1 mm from the sensilla from which the recording was to be made. This wire was used as the reference electrode. The recording electrode was a 250 μm silver/silver chloride wire, placed inside a borosilicate glass capillary (1x0.58x100 mm (ODxIDxL); Harvard apparatus, Holliston, MA, U.S.A.) pulled on a P-97 micropipette puller (Sutter Instrument Co., Novato, CA, U.S.A.) to a tip diameter of ~20-50 μm to fit comfortably over A-type gustatory sensilla on the galea (n=124) without deflecting the hair. Capillaries were filled with tastant solutions (see below) with no added electrolytes. The capillary was moved close to the galea using an LBM-7 manipulator (Scientifica, Uckfield, U.K.) and contact with the apical pore was made using an MO-203 micromanipulator (Narishige, Tokyo, Japan), mounted on the LBM-7.

Electrical signals, which represent the ensemble activity of the four gustatory receptor neurons (GRNs) innervating each sensillum (figure 1a), were recorded through the tastant solutions. The electrodes were connected to a TasteProbe [28] headstage and TasteProbe DTP-02 10x amplifier (Syntech, Buchenbach, Germany). Recordings were made for 20 s with a 1 Hz high-pass filter. The signal was further amplified with 5x gain and band-pass filtered between 10 and 10k Hz (LHBF-48X filter-amplifier, npi electronic GmbH, Tamm, Germany). Recordings were digitised via a CED micro1401-3 digital-to-analogue converter (Cambridge Electronic Design Ltd, Cambridge, U.K.) and acquired using CED Spike2 software (10.09) at 20k Hz.

Gustatory sensilla on the galeae (n=3-6 sensilla per bee) were stimulated with aqueous solutions of D-(+)-sucrose (Fisher Scientific, Loughborough, U.K.), L-valine (Thermo Fisher Scientific, Heysham, U.K.), or L-lysine (Thermo Fisher Scientific, Waltham, MA, U.S.A.), with no added electrolytes. For solutions with valine concentrations above 100 mM, hot (between 60 and 90°C) water was used to aid valine dissolution. Syringes filled with tastant solutions were prepared in advance and stored at -10 °C. Prior to the electrophysiological experiments, the syringes were thawed, and stored at 4 °C. The capillary used for stimulation was filled immediately prior to use, to minimise water evaporation affecting solution concentration. Bees were allocated to one of two experimental groups, one stimulated with low amino acid concentrations and one with high concentrations (figure 1b), to prevent slow adaptation over the course of the experiment. In both conditions, gustatory sensilla were tested initially with 5 mM sucrose and water, to ensure that sucrose sensitivity was comparable between GRN ensembles in all conditions. Bees in the low-concentration group were exposed to increasing concentrations of either lysine or valine from 0.1 to 50 mM. Then, bees in this group were tested with increasing concentrations of sucrose from 1 to 50 mM [27]. The high-concentration group was exposed to increasing concentrations of either lysine or valine from 50 to 500 mM. At least 3 minutes elapsed between exposure to different tastants and concentrations to reduce possible adaptation. Temperature was monitored every 30 s throughout the recording, using an automated temperature logger (EasyLog-USB-2-LCD, Lascar Electronics, Whiteparish, U.K.).

Individual recordings were imported into Matlab (R2024a, MathWorks inc., Natick, MA, U.S.A.) for offline analysis based on [29]. Recordings were trimmed either using contact artefacts, or by the 20 s timestamp acquired from the amplifier, whichever was shortest; the entire trimmed length was used for data extraction. Filtered recordings were then created using a band-pass second-order Butterworth filter between 100 and 1000 Hz. Using these copies, a threshold was then selected manually for each recording to identify spikes. The peak of the spikes above the threshold was used to acquire 4 ms waveforms from the raw signal. The waveforms in each recording were inspected. Four measures were used to distinguish spikes and artefacts: maximum voltage reached, minimum voltage reached, waveform amplitude, and waveform half-width. Individual waveforms with unusual or irregular shapes were considered movement artefacts and removed. Spike frequency was calculated as the total number of spikes in the recording divided by the duration after trimming.