ABSTRACT
After pinealectomy, young pied flycatchers tested in the geomagnetic field have been found to be disoriented. In order to examine the possible role of the pineal hormone melatonin, handraised flycatchers were pinealectomized (PX) at the age of 8 weeks. From the day of operation onward, the PXMEL group received 100 μg of melatonin every evening 1 h before darkness, the PXSOL group was injected with the solvent only, and the PX group was untreated. Unoperated birds served as controls. During the following autumn migration, the birds were tested for directional preference in the local geomagnetic field, in the absence of visual cues. The controls were oriented in the species-specific southwesterly direction; pinealectomized birds without additional melatonin (PXSOL, PX) did not show directional preferences. The PXMEL birds that had received daily injections of melatonin also showed significant southwesterly tendencies; their orientation did not differ from that of the controls. This indicates that melatonin is involved in migratory orientation, either in the processes of expressing the genetically encoded information on the migratory course as a direction with respect to the geomagnetic field or in the time programme controlling the specific migratory direction at a given time.
INTRODUCTION
The amount and the direction of locomotor activity of caged migratory birds during migratory restlessness is closely related to the actual migration of free-living conspecifics (Berthold, 1972; Gwinner and Wiltschko, 1978; Helbig et al. 1989). Displacement experiments and tests with handraised birds indicate that inexperienced birds possess genetically encoded information on the direction and the distance of their migratory route (Perdeck, 1958; Helbig, 1991). These observations led to the hypothesis that an endogenous circannual programme controls the duration of migration and an innate compass course its direction (e.g. Berthold, 1988). Before starting on its first migration, a young bird must transfer the genetically coded directional information into an actual flying direction, which must be maintained over the required distance. The use of the directional information requires an external reference system, which can be provided by the geomagnetic field (Wiltschko and Gwinner, 1974; Beck and Wiltschko, 1982).
Earlier experiments indicated a crucial role of the pineal gland in orientation during the first migration: when pinealectomized at 8 weeks old, pied flycatchers, Ficedula hypoleuca, were unable to orient in the earth’s magnetic field (Semm et al. 1984). This suggested that the pineal gland was involved in the processes of establishing the migratory direction and/or locating it with respect to the magnetic field. The specific role of the pineal gland, however, was still unclear. Pinealectomy is known to result in elimination of the circadian rhythms of locomotor activity and body temperature in sparrows and other passerines (Gaston and Menaker, 1968; Binkley et al. 1971; Menaker and Zimmerman, 1976; Zimmerman and Menaker, 1979; Gwinner, 1981). These effects depend on the absence of the rhythmic synthesis and secretion of the pineal hormone melatonin (Underwood and Goldman, 1987), as they can be cancelled in pinealectomized individuals by the regular application of melatonin (Gwinner and Benzinger, 1978). In connection with migratory orientation, however, mediation of the effects of the pineal gland, either by its neuronal output (Korf et al. 1982; Semm and Demaine, 1984) or by its intrinsic magnetic sensitivity (Demaine and Semm, 1985), seemed equally possible. It was decided, therefore, to examine whether daily injections of melatonin in young, pinealectomized flycatchers could compensate for the effects of pinealectomy on migratory orientation.
MATERIALS AND METHODS
Experimental birds
Pied flycatchers are nocturnal migrants breeding in central Europe and overwintering in western Africa at latitudes between 5° and 12° N. In autumn, the birds migrate first in a southwesterly direction, then change to a southerly or southeasterly direction after they reach the Iberian peninsula, to head for their African winter quarters. During spring migration, the birds fly in a northerly direction using a more direct route via the Sahara and the Mediterranean (Zink, 1985).
Young flycatchers were removed from their nest boxes at approximately 10 days old. They were handraised indoors in the local geomagnetic field. After becoming independent at approximately 4–5 weeks of age, the birds were placed in individual cages in a room with an artificial photoperiod simulating the natural photoperiod of Frankfurt am Main. The flycatchers were thus prevented from experiencing celestial cues.
Experimental treatment
At 8 weeks of age, 24 of the 39 birds were pinealectomized. The operation was conducted under Ketanest (0.4 ml kg−1) and Rompun (0.5 ml kg−1) anaesthesia; 0.02 ml was injected into the pectoralis muscles. Using a pair of extra-fine forceps and a stereomicroscope, the pineal gland and the choroid plexus were carefully removed through a hole that had been drilled in the skull. After the orientation tests the pinealectomy was verified histologically; no remaining pineal tissue was found in any of the birds.
Four groups of birds were used: C, 15 intact birds that served as controls; PX, 6 pinealectomized birds, left untreated; PXMEL 14 pinealectomized birds that received daily melatonin injections; and PX SOL 5 pinealectomized birds that received the solvent only.
For the injections, 50 mg of melatonin (Sigma) was dissolved in 1 ml of absolute ethanol and 9 ml of physiological saline. After the operation, the birds in the PX MEL group received 0.02 ml of the solution, i.e. 100 μg of melatonin daily, a dose that has been shown to be sufficient to minimize the effects of pinealectomy in birds (Underwood and Goldman, 1987). The solution was injected into the right pectoralis muscle, 1 h before the lights were turned off. The PXSOL group received an injection of the solvent only at the same time. This procedure was followed to mimic the natural circadian rhythm of the synthesis and secretion of melatonin (Underwood and Goldman, 1987).
Orientation tests
Orientation experiments were performed from 10 August until 30 September in 1989 and 1990. They took place in small wooden huts in the garden in the local geomagnetic field in Frankfurt (46 000 nT, 66° inclination), in the absence of visual orientation cues. The birds were tested individually in funnel cages (Emlen and Emlen, 1966). For recording, the funnel wall was lined with typewriter correction paper (Rabøl, 1979; Beck and Wiltschko, 1982), upon which the bird’s hopping left scratches. The tests began 30 min after nightfall and lasted about 90 min. All birds were tested only once per night.
Data analysis and statistics
The lining of the funnel cages was divided into 24 sectors of 15° and the number of scratches per sector was counted. Using vector addition, the compass heading was calculated from these data. If the recorded activity was less than 25 scratches, the recording was excluded from analysis.
From the headings of each individual bird, we calculated its mean vector with the heading αb (degrees) and the length rb. Further analysis was based on the mean headings αb. For each test group, a mean vector was calculated and tested for significant directional preference by the Rayleigh test (Batschelet, 1981). The Mardia–Watson–Wheeler test (Batschelet, 1981) was used to compare the distribution of the birds’ headings.
RESULTS
The vectors of the test birds are listed in Tables 1–3. The controls (Table 1) showed significant directional tendencies in a southwesterly direction, which correspond well with the species-specific orientation of free-flying flycatchers, as indicated by ringing recoveries (Zink, 1985). The PX birds, as found in earlier studies, did not show significant preferences, nor did the PXSOL group (Table 2). The PXMEL birds (Table 3), in contrast, were well oriented (see Fig. 1); they are significantly different from the PX and PXSOL groups combined (P<0.01, Mardia–Watson–Wheeler test), but do not differ from the controls (P>0.05).
DISCUSSION
Our data clearly show that the effect of the pineal on migratory orientation is mediated by melatonin. Melatonin could compensate for the orientational deficits normally caused by pinealectomy; birds receiving daily injections of melatonin exhibited the directional tendencies usually observed during that time of the year and their orientation did not differ from that of the intact controls.
It is interesting to consider which stage of the migratory orientation process requires melatonin for normal functioning. Migratory orientation, as we observe it in our test cages, is the result of a complex sequence of processes, many aspects of which are not yet fully understood. Young birds on their first journey to the yet-unfamiliar overwintering site rely on inborn information concerning their migratory direction (Helbig, 1991). This genetically coded directional information must be transferred into an actual direction in space, with the help of an external reference system. Our test birds had been prevented from seeing the sky and observing celestial rotation; hence, the only reference available to them was the geomagnetic field (Beck and Wiltschko, 1982). When autumn migration begins, the birds exhibit migratory restlessness and orient in the seasonally appropriate direction. Is melatonin required for transferring the genetically coded information into a directional angle with respect to the magnetic field, or for initiating the migratory phase so that the directional tendencies can be expressed, or is it involved in the perception and processing of magnetic information?
The latter seems rather unlikely. Pinealectomized homing pigeons, released in overcast conditions, i.e. conditions where they are assumed to rely on the magnetic field, showed normal orientation (Maffei et al. 1983; Papi et al. 1985), which suggests that they could use their magnetic compass. Previous experiments with pied flycatchers also indicated that magnetic orientation per se is largely independent of the melatonin levels: control birds tested at noon, when the melatonin level is at a minimum, were as well-oriented as those tested during the night (Thalau and Wiltschko, 1987), and pinealectomized birds showed well-oriented nocturnal activity during spring migration (Semm et al. 1987). Thus, an effect of melatonin on the magnetic sensor itself or on the capacity to use magnetic information as a compass is unlikely.
Because of the general role of melatonin in the circadian system (Vakkuri et al. 1985), its possible effects on the circadian rhythm or even on the annual cycle (Reiter, 1991; Gwinner et al. 1993) have to be considered. All our test birds showed Zugunruhe, the nocturnal restlessness observed in night-migrating birds during the migratory season, although the amount of activity in the pinealectomized birds did not reach the level of intact controls (H. P. Thalau, unpublished observations). However, the specific preferred directions over the course of migration are also controlled by a time programme (Gwinner and Wiltschko, 1978; Helbig et al. 1989; Munro et al. 1993), and it is still unclear whether this programme is identical with the one controlling activity. Hence, it remains a possibility that low melatonin levels have disruptive effects on aspects of the annual cycle.
It is also possible that the role of melatonin is in the transfer of the genetically encoded directional information into a direction with respect to the magnetic field. These processes, which take place over a limited period prior to the onset of migration, are not yet understood, nor do we know where in the brain information on the migratory direction is processed. However, if melatonin is involved in those processes, our present findings suggest new areas for investigation into where the important steps leading to migratory orientation take place. For instance, the presence of melatonin-binding sites might give an indication of the part of the brain involved.
ACKNOWLEDGEMENTS
This study was supported by the Deutsche Forschungsgemeinschaft in the Program SFB 45 ‘Vergleichende Neurobiologie des Verhaltens’, the Stiftung Volkswagenwerk, and a stipend from the Heisenberg Program to P.S.