Ecotoxicology, 14, 431–437, 2005 2005 Springer Science+Business Media, Inc. Manufactured in The Netherlands.
Methyl Parathion Modifies Foraging Behaviour in Honeybees (Apis mellifera) DAVID GUEZ,* SHAO-WU ZHANG AND MANDYAM V. SRINIVASAN
Research School of Biological Sciences, Australian National University, P.O. Box 475, Canberra, ACT, 0200, Australia Accepted 16 April 2004
Abstract. We examined the effects of sublethal doses of an organophosphorus insecticide, Methyl Parathion (MeP), on the foraging behaviour of honeybees (Apis mellifera ligustica) in a flight cage. The results revealed that MeP modified the frequency of visits to a feeding station to which the bees had previously been trained. A dose of 50 ng per animal elicited an increase in the frequency of visits to the feeder, compared to control animals. A dose of 10 ng, on the other hand, led initially to a decrease in the visit frequency, followed by an increase to a level above that of the controls. A hypothesis is presented to account for the way in which MeP affects foraging behaviour. We propose that the behavioural assay presented here could be useful as a preliminary screening test to study sublethal effects of pesticides on foraging performance in honeybees. Keywords: foraging; cholinergic system; organophosphates; honeybee; Apis mellifera
Introduction Sublethal effects of pesticides on behaviour are difficult to study or predict. Nevertheless, such studies are vital to the development of improved methods of risk assessment and to the evaluation of the ecological and economical impact of insecticides and pesticides in general. It is therefore important to study the potential effects of such molecules on behaviours that are relevant to the organism under study. In the honeybee, one such behaviour is foraging behaviour. This behaviour is crucial not only to the survival and growth of the colony, but also to the commercial value of beekeeping, as it is a prime determinant of honey production. *To whom correspondence should be addressed: Tel.: +61-2-6125-9701; Fax: +61-2-6125-3808; E-mail:
[email protected]
Under normal conditions, the task that a bee performs in its colony tends to be a function of its age (Johnson, 2003). The early phase of a bee’s life is devoted to cleaning the hive and rearing brood. The middle phase involves construction work and guard duty at the hive entrance. The final phase of a bee’s life is devoted to foraging (Huang and Robinson, 1992). An abrupt disappearance of the older bees in a colony – the foragers – either through a natural calamity, or through manipulation by an experimenter, can induce the younger bees to take up foraging prematurely, in order to meet the colony’s foraging needs. Since such ‘‘precocious’’ foragers have had little opportunity or time to carry out the so-called ‘‘orientation’’ flights, where they learn about the outdoor environment around the hive (Capaldi, et al., 2000), many of them can, and do, get lost and are unable to return home. The loss of older bees as well as precocious
432 Guez et al. foragers in this way can have a dramatic impact on the viability of the colony. Therefore, a decrease in the forager population and/or foraging rate that is induced by, say, pesticides in the environment can have dire consequences for the survival of the colony. This would be true not only for lethal doses of pesticide, but also for sublethal doses that modify foraging behaviour in subtle ways. In this study, we investigate the effect of a pesticide on honeybee behaviour by using a simple measure – the foraging frequency of individually marked bees. Measurement of foraging rate is a good first step in assessing the effect of a pesticide, since this parameter is likely to be affected by almost any modification of foraging behaviour. If foraging rate is indeed affected, then subsequent studies can be aimed at uncovering precisely what behavioural modification has led to the change in foraging rate. The need for rapid, preliminary tests of this nature for screening pesticides has been highlighted recently by Thompson (2003). In this study, we use Methyl Parathion (MeP), an insecticide of the organophosphate family, which is known to affect the insect cholinergic pathway by blocking acetylcholinesterase. Another motivation for using MeP is that this substance is often employed as a surrogate to investigate the effect of certain neurotoxins on a variety of organisms. There is therefore the attractive possibility of using the honeybee as a biosensor of potential environmental hazards.
Hive
Materials and methods
7m
We used a four frame nucleus hive of Apis mellifera ligustica. The hive was placed in an insect proof flight cage (10 m · 3 m) with a sucrose feeder (50% v/v sucrose solution) positioned 7 m away from the hive entrance (Fig. 1). A group of bees was trained to visit the feeder. Twenty of the bees were marked individually, using dots of nontoxic acrylic paint. The marked bees were subdivided into two groups of approximately 10 each. One group functioned as a control and the other was treated with MeP during a 10 min window prior to the beginning of the experiments. Bees were treated by topical application of 1 ll DMSO, or 1 ll MeP in DMSO (10 ng/bee, or 50 ng/bee) on the dorsal thorax immediately prior to the experiment. Visits of the marked bees to the feeder were then monitored for 1 hour. The results are displayed as numbers of visits during three successive 20-minute intervals (0–20, 20–40 and 40–60 min). To gather sufficient data the experiment was repeated at least 3 times, using a fresh set of bees each time. The figures show data pooled across all repetitions. All experiments were performed between 13:00 and 14:00 each day during the Austral summer. In a separate experiment, we checked for any possible effects of the solvent DMSO on its own, by using a procedure identical to that described above, to compare foraging rates of bees that had
ete
r Feeder
Figure 1. Experimental design (open view).
MeP Modifies Honeybee Foraging 433 MeP: 88.2% (0–20 min), 94.1% (20–40 min), 97.1% (40–60 min); 10 ng MeP: 82.3% (0–20 min), 94.1% (20–40 and 40–60 min); DMSO: 100% (0– 20 min), 87% (20–40 min) and 91.3% (40–60 min). This indicates that no bee ceased foraging altogether, even if the assiduity of individual bees varied with time, as was the case even for the control bees: 90.4% (0–20 min), 89.4% (20– 40 min), and 89.4% (40–60 min). In the first series of experiments we examined whether the solvent we were using, DMSO, elicited any significant change in foraging frequency on its own. This was done by comparing visit frequencies of DSMO-treated bees with visit frequencies of bees that had received no DSMO, as described in ‘‘Materials and methods’’. Visits of the treated and the control bees were measured over a one-hour period, in three successive 20-minute intervals. The results are shown in Fig. 2. For each interval, there is no significant difference in visit frequencies between the treated and the control bees (p>0.05 in all cases). This finding is further confirmed by sample data for individual bees (Table 1) which shows that the visit rates of individual DMSO treated bees are stable across the different time intervals considered in Fig. 2, and are comparable to those of the untreated bees. Thus DMSO, on its
received 1 ll DMSO, with those of bees that had received no DMSO. All treatments were applied to foraging bees visiting the experimental feeder prior to the beginning of the experiments, when they were imbibing sugar solution and were relatively immobile. Anaesthetic procedures were unnecessary and were not employed. Statistical analysis The data were analysed using the statistical package Systat 10.2. The t-test was used to examine whether differences between the visit frequencies of controls and treated bees were statistically significant, with p £ 0.05 being considered as the threshold for significance.
Results A first important observation is that, regardless of the treatment, all (100%) of the treated bees visited the feeder at least once during the experiments. The proportions of treated bees visiting the feeder during each of the three time intervals under the various treatment regimes were as follows. 50 ng
DMSO Control
n=21 40-60 min
n=20
n=20 20-40 min
n=18
n=23 0-20 min
n=24 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Number Of Visit/20 min Figure 2. Comparison of foraging frequencies of DMSO treated bees and control bees. The data show the means of the numbers of visits of marked bees in the two groups during three successive 20-minute intervals (0–20, 20–40 and 40–60 min). Error bars represent ±SE.
434 Guez et al. Table 1. Number of visits of individual DMSO treated bees and control bees during successive 20-minute periods of the experiment of Fig. 2 and number of visits of individual Mep treated bees and control bees during successive 20-minute periods of the experiment of Fig. 3. DMSO
Control
0–20 min
20–40 min
40–60 min
Bee 1 Bee 2 Bee 3 Bee 4 Bee 5 Mean
3 3 5 3 2 3.2
4 2 4 3 2 3
Bee Bee Bee Bee Bee
Bee 1 Bee 2 Bee 3 Bee 4 Bee 5 Mean
3 2 1 3 2 2.2
5 4 3 6 6 4.8
Bee 1 Bee 2 Bee 3 Bee 4 Bee 5 Mean
5 4 5 4 3 4.2
3 4 4 3 1 3 10 ng/bee 4 3 1 5 4 3.4 50 ng/bee 5 4 5 5 7 5.2
6 5 6 5 7 5.8
own, does not affect foraging frequency, and we conclude that it could be used legitimately as a neutral solvent for administering MeP in the next two experiments, described below. We then examined the effect of MeP on foraging frequency. We investigated two doses, 10 and 50 ng/bee, in separate experiments, as described in ‘‘Materials and methods’’. Since the LD 50 of MeP for summer honeybees is 290 ng/bee (Dr. Jerry Bromenshenk, personal communication) these doses were lower than the LD 50 level by factors of 29 and 6, respectively. In preliminary trials, we did not observe any mortality following topical application of 50 ng/bee of MeP in DMSO, over a subsequent two-day observation period. We first investigated the effect of MeP using a dose of 10 ng/bee (Fig. 3a). The results reveal that, during the first 20 min following the treatment, the MeP treated bees visited the feeder only two-thirds as often as did the control bees. This difference was statistically significant. In the subsequent 20-minute period, the treatment and control groups exhibited similar visit frequencies. In the third 20-minute period, however, the MeP treated bees visited significantly more often than did the control bees.
0–20 min
20–40 min
40–60 min
a b c d e
3 2 4 3 3 3
3 3 5 3 3 3.4
Bee Bee Bee Bee Bee
a b c d e
4 5 4 4 4 4.2
Bee Bee Bee Bee Bee
a b c d e
4 4 3 5 4 4
2 3 4 2 5 3.2 Control 4 4 4 2 4 3.6 Control 4 5 3 4 4 4
4 5 4 3 3 3.8 4 4 4 3 5 4
Clearly, MeP affects foraging frequency, and the effect varies with time, being inhibitory initially and facilitatory at later times. We then examined the effect of a stronger dose of MeP, namely 50 ng/bee. The results are shown in Fig. 3b. During the first 20 min after treatment, the MeP treated bees and the control bees visited the feeder with similar frequencies. However, during the subsequent two 20-minute periods the MeP treated bees foraged significantly more often than did the controls. Thus, whereas the 50 ng dose induced a consistent increase in foraging rate, the lower dose (10 ng) had a more complex effect, causing an initial decrease in foraging frequency, followed by a subsequent increase. The sample data for individual bees in this experiment (Table 2) reveal that the performance of individual bees mirrored the average trends observed in Fig 3, for the two MeP dosages. Thus, it is clear that MeP affected foraging behaviour on an individual basis, and that the changes in the observed foraging rate were not caused by some individuals foraging at a given rate during one time period and other individuals foraging at a different rate during another time period. The 10 ng dose of MeP had an immediate effect on foraging, inducing a clear
MeP Modifies Honeybee Foraging 435 (a)
Methyl Parathion 10 ng/bee Control
40-60 min
n=15
*
n=23 n=16 20-40 min
n=27
0-20 min
n=14
*
n=23
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 Methyl Parathion 50 ng/bee Control
(b)
40-60 min
n=33
*
n=41
20-40 min
n=31
*
n=41 n=29 0-20 min
n=39 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Number of Visit/20 min Figure 3. Comparison of foraging frequencies of MeP treated bees and control bees. The data show the means of the numbers of visits of marked bees in the two groups during three successive 20-minute intervals (0–20, 20–40 and 40–60 min). (a) Results for MeP dosage of 10 ng/bee; (b) results for MeP dosage of 50 ng/bee; *denotes statistically significant difference at p < 0.05. Error bars represent ±SE. In these experiments, there are fewer treated bees than controls because the number of treated bees was limited by the number of treatment-designated bees that actually arrived at the feeder during the 20-minute treatment period (see Materials and methods).
decrease in foraging rate over the first 20 min, whereas the stronger dose had no effect at all during the first 20 min period, and enhanced foraging only during the two subsequent periods. As we shall discuss below, these qualitative, dose-dependent differences in the foraging patterns may provide some insights into the mechanisms by which MeP influences foraging behaviour.
Discussion Our findings reveal that MeP modifies foraging behaviour, resulting in a change in the rate at which a food source is visited. The higher dose (50 ng/bee) induces a progressive increase in foraging rate, whereas the lower dose elicits an early decrease in foraging rate, followed by an increase.
436 Guez et al. Methyl Parathion is known to block acetylcholinesterase activity, and, therefore, to increase cholinergic transmission along the nicotinic as well as the muscarinic pathways. It is therefore very likely that the observed behavioural changes are caused by changes in the level of cholinergic transmission. The results reveal, however, that MeP affects foraging behaviour in a rather complex way, producing qualitative as well as quantitative changes. We advance the following hypothesis to account for the observed changes in foraging rate. First, we assume that the site at which MeP acts to modify the observed behaviour is some distance away from the site at which it is applied (the surface of the thorax). This is a reasonable assumption, as the brain is the most likely site at which such complex changes in behaviour are likely to be mediated. This assumption implies that there is a time delay between the application of the drug and the time at which the concentration of the drug reaches a concentration that is high enough to have an effect. Second, we assume that low concentrations of MeP at the site of action (or, equivalently, small increases of cholinergic transmission) lead to a decrease in foraging frequency. Concentrations greater than a critical level, however, lead to an increase in foraging frequency. Given these two assumptions, the response patterns of the bees to the low and high dosages of MeP can be explained as follows. At the low dosage (10 ng), the concentration of MeP at the site of action is initially low for a considerable time before it exceeds the critical level. This causes the foraging frequency to decrease initially, and increase at later times. At the high dosage (50 ng), the MeP concentration at the site of action increases so rapidly that it exceeds the critical level almost immediately after application. As a result, one observes only an increase in foraging frequency. During the initial part of the first 20minute period the MeP concentration is below the critical level; during the latter part it is above the critical level. Consequently, when the number of visits is counted over the entire 20-minute period, as we have done, there is no apparent change in foraging frequency. The above hypothesis can be tested rigorously by using a larger number of concentrations, and by measuring the changes in foraging frequency with finer temporal resolution.
It is also important to point out that this hypothesis requires only one target of action for the MeP (namely AchE) , as the behavioural effects are postulated to be mediated by variations in the level of acetylcholine. What is the specific nature of the behavioural change that causes the foraging frequency to increase at high dosages of MeP? Our data do not provide an answer to this question. However, our preliminary observations rule out a few possibilities. First, MeP treated bees stay at the feeding station for about the same duration as the control bees. Second, MeP treated bees do not seem to spend longer outside the hive during each foraging bout (for example, resting on the floor or walls of the cage) than the control bees. Schricker and Stephen (1970) demonstrated that Parathion treatment led to a modest (10%) increase in flight speed. However, such a change would be too small to account for the observed increase in foraging frequency in our experiments, because flight to and from the feeder constitutes a very small fraction of the total duration of a foraging trip. Given the small distance between the hive and the feeder in our experiments (7 m), most of the duration of a foraging cycle is spent at the feeder, collecting the food, and in the hive, offloading the food to nestmates, and, possibly, performing a dance to recruit other foragers to the feeder. In our preliminary observations of activity in the hive, done in an open field with a feeder placed 150 m from the hive entrance, we found that, with the high dose of MeP, the treated bees spent a noticeably shorter time in the hive, compared to the control bees. Indeed, the treated bees rarely danced: they appeared to be ‘‘restless’’ and often displayed the so-called ‘‘trembling’’ movements that are known to play a role in attracting nestmates to offload the food as quickly as possible (Von Frisch, 1993; Seeley et al., 1996; Dyer, 2002). Furthermore, the treated bees rarely ventured deep into the hive: they tended to unload their food very close to the entrance, before flying out on the next foraging bout. Thus, the principal reason for the increase in foraging rate that is elicited by high dosages of MeP appears to be the reduced time spent in the hive, which in turn is likely to be due to (a) a reduced tendency to perform dances, (b) attracting
MeP Modifies Honeybee Foraging 437 more nestmates to offload the food rapidly, and (c) offloading the food very close to the entrance. Obviously, further more detailed study is required to confirm our observations at the hive. Nevertheless, the present investigation demonstrates that MeP clearly affects foraging behaviour in honeybees; and that measurement of foraging rate offers a rapid and convenient means of screening pesticides for their potential effects on the behaviour of social insects.
Acknowledgements This research was partially supported by Grant N00014-99-1-0506 from the U.S. Defence Advanced Research Projects Agency and the Office of Naval Research to MVS. David Guez was partly supported by a Bourse Lavoisier. We thank David Bromenshenk for information on the LD 50 of Methyl Parathion for bees, and Alan Rudolph and Joel Davis for advice and encouragement.
References Capaldi, E.A., Smith, A.D., Osborne, J.L., Fahrbach, S.E., Farris, S.M., Reynolds, D.R., Edwards, A.S., Martin, A., Robinson, G.E., Poppy, G.M. and Riley, J.R. (2000). Ontogeny of orientation flight in the honeybee revealed by harmonic radar. Nature 403, 537–540. Dyer, F.C. (2002). The biology of the dance language. Ann. Rev. Entomol. 47, 917–949. Huang, Z.Y. and Robinson, G.E. (1992). Honeybee colony integration: worker–worker interactions mediate hormonally regulated plasticity in division of labor. Proc. Natl. Acad. Sci. USA 89, 11726–11729. Johnson, B.R. (2003). Organization of work in the honeybee: a compromise between division of labour and behavioural flexibility. Proc. Roy. Soc. Lond. B Biol. Sci. 270, 147–152. Schricker, B. and Stephen, W.P. (1970). The effect of sublethal doses of parathion on honeybee behaviour. I. Oral administration and communication dance. J. Apicult. Res. 9, 141– 153. Thompson, H.M. (2003). Behavioural effects of pesticides in bees – Their potential for use in risk assesment. Ecotoxicology 12, 317–330. Seeley, T.D. (1996). The Honey bee’s tremble dance stimulates additional bees to function as nectar receivers. Behav. Ecol. Sociobiol. 39, 419–427. Von Frisch, K. (1993). The Dance Language and Orientation of Bees. Harvard University Press, Cambridge, Massachusetts London, England.
