Honey bee circadian clocks: behavioral control from individual workers to whole-colony rhythms

Abstract

In the field of insect circadian rhythms, the honey bee is best known for its foraging time-sense, or Zeitgedächtnis, which permits the forager bee to make precise associations between the presence of food and the time of day. A number of studies, now considered classics, established that bees could be trained to collect food at virtually any time of the circadian cycle and that this timekeeping ability was controlled by an endogenous circadian clock. Recently, behavioral rhythms in bees have been examined using a variety of approaches, in both laboratory and field studies. The following areas of new research are reviewed: (a) the ontogeny of behavioral rhythmicity in newly emerged worker bees; (b) the integration of behavioral rhythmicity with the colony's division of labor; (c) the evidence for social entrainment of behavioral rhythms and for a ‘clock of the colony’; (d) the potential linkage between circadian rhythms of general locomotor activity and the foraging time-sense; (e) learning and entrainment hypotheses proposed to explain the mechanism underlying the time-sense; (f) the interplay between extinction and persistence of the time-memory as revealed from the differential behavior of individuals within the foraging group; and (g) comparisons of the Zeitgedächtnis with food-anticipatory rhythms in other animals.

Introduction

It has been known since the turn of the last century that honey bees (Apis mellifera) are able to learn the time of day when flowers secrete nectar (von Buttel-Reepen, 1900). Forager bees return to a food source at the same time on consecutive days and this capability persists for several days after the food source is removed (Forel, 1910). This apparent memory for time, or Zeitgedächtnis, was tested experimentally by Beling (1929) and Wahl (1932) who showed that the bees could be trained to collect nectar and pollen at virtually any time of day.

The early work on the honey bee foraging rhythm is considered to be a historical milestone, providing the first convincing evidence that circadian systems permit organisms to measure time for adaptively significant purposes (Moore-Ede et al., 1982). Bees synchronize their behavior with daily floral rhythms, foraging only when nectar and pollen are at their highest levels. At other times, they remain in the hive, conserving energy that otherwise would be exhausted on non-productive foraging flights (Kleber, 1935). The resting bees may even station themselves in relatively remote locations in the hive, distant from the intense activity of the dance floor (Körner, 1939, von Frisch, 1940, Moore et al., 1989). Because the foraging behavior free-runs under constant conditions with a periodicity deviating slightly from 24 h, the Zeitgedächtnis is assumed to be driven by an endogenous circadian oscillator (Renner, 1955, Renner, 1957, Beier, 1968, Beier and Lindauer, 1970, Frisch and Aschoff, 1987). Consistent with other circadian phenomena, the foraging rhythm is entrained by light–dark (LD) cycles and may be phase-shifted, with transients, to phase changes of the LD cycle (Renner, 1959, Beier, 1968). Like other circadian rhythms, the foraging rhythm has a relatively narrow range of entrainment and will fail if LD cycles are shorter than 20 h or longer than 26 h (Beier, 1968).

In all the foraging rhythm studies, the temporal response is measured from the distribution of arrivals of many individuals at the training station. The foraging rhythm is therefore a population response, derived from the activity of many individuals comprising a foraging group. Behavioral rhythms also have been measured in isolated individual bees. In accord with many other insect rhythms (Aschoff, 1979), the free-running period of locomotor activity in individual foragers is shorter in DD than in LL (Moore and Rankin, 1985), contravening Aschoff's (1960) ‘circadian rule’ that diurnal animals should decrease circadian period (i.e. increase frequency) with changes from DD to LL. On the other hand, the bee locomotor rhythm does follow the generalization that for species showing τ_DD<24 h, τ is lengthened in LL (Daan and Pittendrigh, 1976). At the cellular level, preliminary results (Bloch et al., 1999) show that bees, in common with many other insects, possess PER-immunoreactive cells in the lateral protocerebrum of the brain, and that the levels of PER in these cells oscillate in a circadian fashion, peaking at night.

Controversial studies conducted in the 1970s and 1980s challenged the widely held view that the honey bee foraging rhythm was based on an endogenous circadian clock (Martin et al., 1978, Martin et al., 1983, Martin and Martin, 1987). The authors claimed that the learned foraging time is simply cued to local diel changes in the earth's magnetic field. According to their hypothesis, bees form a memory of local geomagnetic field conditions that are present when they receive food rewards and these memories are stored within the mushroom bodies (corpora pedunculata) of the brain. Further, the time-memory could be transferred between individuals if the mushroom bodies from donor bees were transplanted into the head capsules of recipient foragers. A reanalysis of the original data (Brady, 1987), however, in addition to pointing out several methodological problems in the original studies, has shown than the results are not inconsistent with the endogenous interpretation. Providing convincing evidence that honey bee circadian rhythms are generated endogenously rather than exogenously are long-term, free-running rhythms under conditions in which local changes in the magnetic field are not controlled. Such free-runs have been demonstrated for general locomotor (walking) behavior in individual bees (Moore and Rankin, 1985, Toma et al., 2000) as well as for foraging behavior in whole colonies (Frisch and Aschoff, 1987). In a direct experimental test of the exogenous hypothesis, Neumann (1988) compared trained bees in natural and altered magnetic fields and found no significant differences in the timing of their foraging behavior. Thus, the contention that honey bee circadian rhythms are entirely exogenous in origin (Martin and Martin, 1987) appears to have been dispelled.

Although the honey bee Zeitgedächtnis shares many attributes with typical circadian rhythms, it also has some extraordinary properties. Because of its unique characteristics, Saunders (1982), in his classic review of insect circadian rhythms, places the honey bee foraging rhythm in a chapter entitled “Other Types of Insect Clock”. Unlike typical circadian rhythms, in which the behavior or physiological process is expressed only at a particular phase position within the cycle, the Zeitgedächtnis gives the honey bee forager the ability to recognize any time of day and adjust its behavior accordingly. This versatility in time-keeping ability suggests a ‘continuously consulted’ circadian oscillator (Pittendrigh, 1958). A remarkable example of continuous clock consultation is provided by ‘marathon dancers’. These are foragers that perform the recruitment waggle dance for hours at a time, without leaving the dark confines of the hive, accurately indicating the direction of a food source with respect to the sun's azimuth at any time of the day or night. The honey bee Zeitgedächtnis also may involve a sophisticated form of learning, as a wide variety of information is stored in a time-linked fashion, including spatial location (Wahl, 1932), nectar concentration (Wahl, 1933), visual and olfactory cues (Koltermann, 1974), and even the specific body orientation of approach for flower-landing (Gould, 1987). Thus, it appears that all of the sensory signals associated with flower location and recognition are linked together with time to form a learned Gestalt; altering one signal causes the bee's orientation to the entire configuration to deteriorate (Bogdany, 1978).

Despite the long tradition of work in this field, many fundamental questions concerning the circadian control of behavior in honey bees remain unanswered. The major goal of this review is to examine some of those outstanding questions and to present the results of recent and current research designed to address some of the gaps in our knowledge. Because the honey bee is not a solitary organism but rather lives in a complex, highly organized society often consisting of tens of thousands of individuals, particular attention is given to those studies which focus on the integration of circadian rhythms in the individual with the colony's social structure, especially the age-related division of labor.