RFID tracking of sublethal effects of two neonicotinoid insecticides on the foraging behavior of Apis mellifera

Abstract

The development of insecticides requires valid risk assessment procedures to avoid causing harm to beneficial insects and especially to pollinators such as the honeybee Apis mellifera. In addition to testing according to current guidelines designed to detect bee mortality, tests are needed to determine possible sublethal effects interfering with the animal's vitality and behavioral performance. Several methods have been used to detect sublethal effects of different insecticides under laboratory conditions using olfactory conditioning. Furthermore, studies have been conducted on the influence insecticides have on foraging activity and homing ability which require time-consuming visual observation. We tested an experimental design using the radiofrequency identification (RFID) method to monitor the influence of sublethal doses of insecticides on individual honeybee foragers on an automated basis. With electronic readers positioned at the hive entrance and at an artificial food source, we obtained quantifiable data on honeybee foraging behavior. This enabled us to efficiently retrieve detailed information on flight parameters. We compared several groups of bees, fed simultaneously with different dosages of a tested substance. With this experimental approach we monitored the acute effects of sublethal doses of the neonicotinoids imidacloprid (0.15-6 ng/bee) and clothianidin (0.05-2 ng/bee) under field-like circumstances. At field-relevant doses for nectar and pollen no adverse effects were observed for either substance. Both substances led to a significant reduction of foraging activity and to longer foraging flights at doses of ≥0.5 ng/bee (clothianidin) and ≥1.5 ng/bee (imidacloprid) during the first three hours after treatment. This study demonstrates that the RFID-method is an effective way to record short-term alterations in foraging activity after insecticides have been administered once, orally, to individual bees. We contribute further information on the understanding of how honeybees are affected by sublethal doses of insecticides.

Figure 1. Automatic registration at the feeder.

(A) The RFID-transponder attached to the thorax shield of honeybees allows tracking of the foraging activity with RFID-readers positioned at hive entrance and at the feeder. (B) Bees foraging from an artificial feeder placed in a feeder compartment. The bar-shaped scanners at the entrance of the feeder compartment detect every passage of a labeled honeybee, when it passes through specially crafted bee tunnels (see also Figure S4).

Figure 2. Changes in the frequency of feeder visits after treatment with imidacloprid or clothianidin.

During 3-hour observation periods, we recorded the frequency of visits at the feeder immediately after treatment (a.t.) and up to 48 h after treatment (24 h a.t., 48 h a.t.). Values inside the bars: n = Number of bees returning to the feeder after treatment, % = (bees returned/bees treated)*100, (A) Oral administration with 0.15 ng imidacloprid did not affect the visit frequency per bee significantly, while 1.5 and 3 ng imidacloprid led to a significant reduction of feeder visits compared to the controls during the first three hours immediately after treatment. Not all bees treated with 3 ng reappeared at the feeder immediately after treatment, but almost all returned regularly after 24 h. No bees treated with 6 ng imdacloprid returned in the first three hours of observation after treatment. After 24 h only three of twelve bees returned to forage from the feeder. (B) The frequency of visits was not affected negatively after administration of 0.05 ng clothianidin, while treatment with 0.5 ng, 1 ng, and 2 ng clothianidin reduced the frequency of visits significantly compared to the control group during the first three hours immediately after treatment. As shown here, only 67.8% and 11.8% of the bees treated with 1 ng and 2 ng, respectively returned to forage at the feeder. The missing bees were not registered again during the experiments. The significant reductions to the visit frequency by both substances did not persist on the following day. * = p<0.05, ** = p≤0.01, *** = p≤0.001.

Figure 3. Influence on foraging trip duration and its different phases after treatment with imidacloprid.

Plotted were the median times in seconds for every bee during 3-hours observation periods immediately after treatment (a.t.) and 24 h after treatment (24 h a.t.). * = p<0.05, ** = p≤0.01, *** = p≤0.001. (A) Imidacloprid: Bees treated with 1.5 ng and 3 ng imidacloprid spent more time outside of the hive for a foraging trip compared to the control group during the first three hours immediately after treatment. These effects were not persistent after 24 h. (B) We observed significantly prolonged flights to the feeder and (D) back to the hive during the three hour observation period after oral administration, for bees treated with 1.5 and 3 ng imidacloprid compared to the control group during the first three hours immediately after treatment. (C) Bees treated with 1.5 and 3 ng imidacloprid spent more time inside the feeder compartment compared to the control group during the first three hours after treatment. This effect was not found to be persistent 24 hours after administration.

Figure 4. Influence on foraging trip duration and its different phase after treatment with clothianidin.

Plotted were the median times in seconds for every bee during 3-hours observation periods immediately after treatment (a.t.) and 24 h after treatment (24 h a.t.). * = p<0.05, ** = p≤0.01, *** = p≤0.001. (A) After bees were treated with 0.5 ng, 1 ng, and 2 ng clothianidin their median time spent outside of the hive was significantly prolonged compared to the control group. On the following day we observed slightly but significantly shorter foraging trips by the bees treated with 0.05 ng compared to the control. Foraging trips by the bees treated with 2 ng clothianidin were still found to be significantly prolonged after 24 hours, though not as profound compared to the day before. (B) Treatment with clothianidin, regardless of the dose, showed no significant effect on flight time to the feeder. (C) Immediately after treatment with 0.5 ng, 1 ng and 2 ng clothianidin, bees spent more time inside the feeder compartment compared to the control group. Twenty-four hours after treatment we observed significantly shorter feeder visits for bees treated with 0.05 ng when compared to the control, while no significant difference was observed for bees treated with the other doses. (D) Bees treated with 0.5 ng, 1 ng, and 2 ng needed significantly longer to fly back to the hive compared to the controls during the three hour observation period immediately after treatment. After 24 h bees treated with 2 ng still needed significantly longer than the control group when returning to the hive, though the difference was not as profound compared to the day before.

Figure 5. Time interval between foraging trips after treatment with both insecticides.

Plotted was the median period spend inside the hive between two foraging trips and the duration of the first in-hive stay for every bee (in minutes) immediately after respective administration of one of the two insecticides.* = p<0.05, ** = p≤0.01, *** = p≤0.001. (A) In the three hour observation period after treatment with 1.5 ng and 3 ng imidacloprid and (C) 0.5 ng, 1 ng and 2 ng clothianidin, these bees needed significantly longer to fly out again after returning from for the subsequent foraging trip compared to the control groups. (B) Administration of 3 ng imidacloprid led to a significantly prolonged first stay inside of the hive. (D) Bees that were treated with 1 ng and 2 ng clothianidin had longer first in-hive stays compared to the controls.