J Comp Physiol A (2014) 200:399–407 DOI 10.1007/s00359-014-0898-y

Original Paper

Orientation of migratory birds under ultraviolet light Roswitha Wiltschko · Ursula Munro · Hugh Ford · Katrin Stapput · Peter Thalau · Wolfgang Wiltschko 

Received: 31 December 2013 / Revised: 5 March 2014 / Accepted: 7 March 2014 / Published online: 10 April 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract  In view of the finding that cryptochrome 1a, the putative receptor molecule for the avian magnetic compass, is restricted to the ultraviolet single cones in European Robins, we studied the orientation behaviour of robins and Australian Silvereyes under monochromatic ultraviolet (UV) light. At low intensity UV light of 0.3 mW/m2, birds showed normal migratory orientation by their inclination compass, with the directional information originating in radical pair processes in the eye. At 2.8 mW/m2, robins showed an axial preference in the east–west axis, whereas silvereyes preferred an easterly direction. At 5.7 mW/m2, robins changed direction to a north–south axis. When UV light was combined with yellow light, robins showed easterly ‘fixed direction’ responses, which changed to disorientation when their upper beak was locally anaesthetised with xylocaine, indicating that they were controlled by the magnetite-based receptors in the beak. Orientation under UV light thus appears to be similar to that observed under blue, turquoise and green light, albeit the UV responses Electronic supplementary material  The online version of this article (doi:10.1007/s00359-014-0898-y) contains supplementary material, which is available to authorized users. R. Wiltschko (*) · K. Stapput · P. Thalau · W. Wiltschko  Fachbereich Biowissenschaften der, J.W.Goethe-Universität Frankfurt, Max von Laue Straße 13, 60438 Frankfurt am Main, Germany e-mail: [email protected]‑frankfurt.de U. Munro  School of the Environment, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia H. Ford  Division of Zoology, School of Environmental and Rural Sciences, University of New England, Armidale, NSW 2351, Australia

occur at lower light levels, probably because of the greater light sensitivity of the UV cones. The orientation under UV light and green light suggests that at least at the level of the retina, magnetoreception and vision are largely independent of each other. Keywords  Magnetoreception · UV cones · Monochromatic light · Migratory orientation · ‘Fixed direction’ responses

Introduction The primary processes by which birds can detect the direction of the geomagnetic field were unknown until, in 2000, Ritz and colleagues proposed the Radical Pair Model, a mechanism based on spin chemical processes. Subsequent experiments with radio frequency fields, a diagnostic tool for radical pair processes (Ritz 2001; Henbest et al. 2004), supported this model (Ritz et al. 2004, 2009; Thalau et al. 2005; W. Wiltschko et al. 2007; Keary et al. 2009). Magnetoreception was found to take place in the eye (e.g. W. Wiltschko et al. 2002, 2003a; Stapput et al. 2010), with information mediated and processed by part of the visual system (Semm et al. 1984; Semm and Demaine 1986; Heyers et al. 2007; Zapka et al. 2009). Cryptochrome, the only photopigment in animals with the required properties, has been suggested to be the receptor molecule that forms the crucial radical pairs (Ritz et al. 2000). Several cryptochromes—Cry1a, Cry1b, Cry2 and Cry 4—have been found in the eyes of domestic chickens, Gallus gallus, and migratory birds (Bailey et al. 2002; Haque et al. 2002; Möller et al. 2004; Watari et al. 2012; for summary, see Liedvogel and Mouritsen 2010), but only recently, an immuno-histological study revealed details of

13

400

their location in the retina (Nießner et al. 2011): birds have four colour cones; Cry1a occurs in the outer segment of a specific type of photoreceptor, the single cone with SWS1 opsin, where it is associated with the disk membranes. This cone type, which is sensitive to ultraviolet (UV) light in European Robins, Erithacus rubecula (Muscicapidae, Turdinae) and to violet light in domestic chickens, is distributed all across the retina, allowing a comparison of activation in the various directions. Cry1a thus fulfils the requirements of the radial pair model (Ritz et al. 2000), suggesting that it is indeed the receptor molecule for magnetic compass information. The first step in forming the radical pairs is photon absorption, which leads to the wavelength-dependency of avian magnetoreception: it requires light from the shortwavelength range of the spectrum. Experiments with migratory birds such as European Robins, Garden Warblers, Sylvia borin, and Australian Silvereyes, Zosterops lateralis, with homing pigeons, Columba livia f. domestica, and domestic chickens have shown that their magnetic compass works well under blue, turquoise and green light up to 565 nm, whereas birds are normally disoriented under 582 nm yellow light and longer wavelengths (for summary, see Wiltschko et al. 2010). Birds have four spectral cone types, and the vision of most passerines, including the families Muscicapidae and Zosteropidae, extends into the UV range well below 400 nm. In view of the finding that cryptochrome 1a occurs exclusively in the UV cones, the response under short wavelength below 400 nm has now become of special interest: how does visual activation of this cone type affect magnetoreception? Magnetoreception under UV light has not yet been analysed in detail. Here, we report the results of experiments with European Robins and Australian silvereyes. Analogous to earlier experiments under monochromatic blue, turquoise and green light (see R. Wiltschko et al. 2007, 2010), we tested the birds under 373 nm ultraviolet light of varying intensities and in combination with yellow light.

Materials and methods The experiments with European Robins were performed in Frankfurt am Main, Germany (50°08′N, 8°40′E) in the years from 2001 to 2012, those with silvereyes in Armidale, NSW, Australia (30°30′S, 151°40′E) in 2003. Test birds The European Robins tested were juvenile birds of probably Scandinavian origin. They were caught as transmigrants in the Botanical Garden near the Zoological Institute in Frankfurt am Main in early September. From the capture

13

J Comp Physiol A (2014) 200:399–407

onward, they were kept in individual cages under a photoperiod simulating the natural one during the autumn tests, which took place from mid-September to mid-October, until early December, when it was L:D 8:16. Around new year, the duration of the light phase was increased in two steps to L:D 13:11. This induced premature Zugunruhe for spring migration and allowed us to test the birds once more from mid-January to late February. At the end of March, when the natural photoperiod had reached L:D 13:11, the birds were set free again. The Australian Silvereyes were members of the migratory Tasmanian subspecies Zosterops l. lateralis that can be recognised by their slightly larger size, rufous flanks and whitish-grey throats. Caught in early southern spring during the first half of September in Armidale, they were either part of the local wintering population or on their return journey from southern Queensland to Tasmania. Being a social species, they were kept four birds in each cage under a photoperiod that corresponded to the natural one during the spring migration tests, which took place from late September to late October. When these tests were finished, the birds were released into the wild again. Test performance Our standard method for testing migratory birds was used: the birds were tested one at a time in funnel-shaped cages (upper diameter 35 cm, Emlen and Emlen 1966), the inclined walls of which were lined with type-writer correction paper (BIC, Germany, formerly Tipp-Ex).When moving, the bird left scratch marks on the paper lining. Each funnel cage was placed in a cylinder, 50 cm high, with the top of the cylinder consisting of the disk carrying the lightemitting diodes (LEDs). We used UV-LEDs with a peak wavelength of 373 nm (half intensity at wavelengths 368 and 381 nm); in some tests with European Robins, these LEDs were combined with yellow LEDs with a peak wavelength of 585 nm (half intensity 571 and 604 nm). The light intensity was measured before each test with an optometer (Gigahertz-Optik, Puchheim, Gemany). In previous experiments with blue, turquoise and green light, we had used an equal quantal flux of about 8 × 1015 quanta/s m2 as a standard light level (see R. Wiltschko et al. 2010); in the present experiments, we used the same light level, which corresponds to 2.8 mW/m2 in UV. In Frankfurt, this is similar to the level of the UV part of the spectrum on a clear day about 30 min after sunset or before sunrise. The birds were also tested at a lower light level of only 1/10 of that intensity, 0.3 mW/ m2, corresponding to the level of the respective part of the spectrum observed more than 1 h after sunset or before sunrise. Robins were additionally tested at a light level twice the standard, 5.7 mW/m2. The 585 nm yellow light

J Comp Physiol A (2014) 200:399–407

that was added to the UV light also had the standard quantal flux, corresponding to 1.7 mW/m2. For control, we used light conditions under which the birds had been shown to be well oriented in their migratory direction: “white” light produced by a 15 W light bulb, turquoise light with a peak wavelength of 502 nm or green light with a peak wavelength of 565 nm, the latter two produced by LEDs (for details, see Table S1 in Electronic supplementary material). The light passed through two diffusers before it reached the bird in the cage. Most tests were performed in the local geomagnetic field, which was about 46 μT, 66° inclination in Frankfurt, Germany, and about 56 μT, −62° inclination in Armidale, Australia. At both sites, birds were additionally tested in a magnetic field with the vertical component inverted by Helmholtz coils. In Frankfurt, robins were also tested with an oscillating field of 1.315 MHz, 480 nT added vertically, i.e. at an angle of 24° to the magnetic vector. It was produced by a coil antenna mounted horizontally so that the axis of the oscillating field was vertical, i.e. at an angle of 24° to the static geomagnetic field (see Ritz et al. 2004; Thalau et al. 2005 for details). Furthermore, the robins were tested with the magnetite-based receptors in the upper beak (see Fleissner et al. 2003; Falkenberg et al. 2010) anesthetized with the local anaesthetic xylocaine 2 % (active substance: Lidocainhydrochlorid 1 H2O, Astra Zeneca GmbH) which was applied externally by gently moving a cotton bud soaked in xylocaine along the skin of the upper beak. Testing began when the light went out in the bird rooms and lasted for about 75 min. Each bird was tested three times in each test condition.

401

into the next step of the analysis. From the three headings of a bird under each condition, we calculated the mean vector of that bird with the heading αb and the length rb. For birds that showed axial behaviour, we calculated the axial vector and chose the end that was closer to the unimodal mean for further analysis. The mean headings or ends of the mean axes αb of all test birds were then combined in the grand mean vector for each condition, with the direction αN and the length rN, and, by doubling the angles, a grand mean axis. The grand mean vectors and the grand mean axes were tested by the Rayleigh test for significant directional preferences (Batschelet 1981). This evaluation procedure was applied to all data, even to the one with only occasional axiality. The data from different test conditions were compared with the Watson Williams test to look for differences in direction (if both rN > 0.65), with the non-parametric Mardia Watson Wheeler test to look for differences in distribution and with the Mann–Whitney test applied to the differences of the birds’ headings αb from the mean to look for differences in variance. We also calculated the medians of the vector lengths per bird, which reflects the intra-individual variance.

Results The tests with European Robins were performed in the northern autumn and spring, and those with Australian Silvereyes in the southern spring.

Data analysis

Orientation of European Robins under monochromatic ultraviolet light

To obtain the birds’ directional preferences, the coated paper was removed, divided into 24 sectors, and the scratch marks in each sector were counted by a person who was blind to the test condition. Recordings with a total of fewer than 35 scratches were excluded from the analysis because of too little activity and the birds were tested again. From the distribution of the activity within the cage, the heading of the respective test was calculated by vector addition. Under brighter UV light, there was an unusually high number of recordings where the bird showed an axial preference, that is, there was a peak of activity in one direction and another one roughly opposite; hence, we applied the procedure described by R. Wiltschko et al. (2007). We calculated the unimodal headings and also the axial headings of each recording by doubling the angles. For further analysis, we used the heading that produced the longer vector; if this was the axis, we entered the end with more activity (i.e. the end that lay closer to the unimodal heading)

When we began testing European Robins under monochromatic UV light, we first used an intensity of 2.8 mW/m2 corresponding to a quantal flux of 8 × 1015 quanta/s m2, a light level at which we had found good orientation in migratory direction under blue, turquoise and green light (see R. Wiltschko et al. 2010). Under UV light of this intensity, however, the robins were not oriented in their migratory direction; instead, they showed a not very pronounced preference for an axis running roughly east–west in both autumn and spring (Fig. 1). Not only was the activity within the cage often axially distributed, but also the fraction of birds with axial vectors, mostly exceeding 50 %, was remarkably high (Table 1). Orientation in the migratory direction, as shown in the control tests, was observed when the light intensity was reduced to 0.3 mW/m2, one tenth of the initial light level. When the intensity of the UV light was doubled to about 5.7 mW/m2 in spring, the robins’ behaviour changed again, with the birds now showing a preference for an axis roughly running north–south (see

13

402

J Comp Physiol A (2014) 200:399–407

Fig. 1  Orientation of European Robins under UV light of different intensities. C control, UV UV light with a peak wavelength of 373 nm; the various intensities are indicated in the diagram. The triangles at the periphery of the circle mark mean headings of individ-

ual birds (open preferred end of a mean axis), the arrows indicate the mean vectors and double arrows mean axes. The two inner circles are the 5 % (dotted) and the 1 % significance border of the Rayleigh test. For numerical values, see Table 1

Table 1  Orientation of robins in UV light of different intensities in the local geomagnetic field Season

Condition

N

Axial record (%)

Axial birds (%)

Med. rb

αN

rN

ΔC

Autumn

Control UV 0.3 UV 2.8 Control UV 0.3 UV 2.8

32 32 32 24 24 24

8 9 27 8 8 31

25 31 56 29 8 54

0.83 0.79 0.92 0.89 0.95 0.74

176° 182° 78°–258° 10° 13° 86°–266°

0.85*** 0.53*** 0.63*** 0.92*** 0.96*** 0.44**

UV 5.7

12

19

75

0.69

13°–193°

0.76***

C +6°n.s. +82°*** C +3°n.s. +76°**

Spring

+1°n.s.

Conditions: UV 0.3 ultraviolet light with a peak at 373 nm and an intensity of 0.3 mW/m2, UV 2.8 and UV 5.7 the same UV light at the indicated intensities, N number of birds tested; the following two columns give the percentage of axial recordings and the percentage of birds producing axial vectors, med rb median of the unimodal or axial vector lengths of the individual birds, αN, rN direction and lengths of the grand mean vector; in case of axes, both ends are indicated, ΔC angular difference to the control direction Asterisks at the vector lengths indicate a significant directional preference (Rayleigh test). Asterisks at ΔC indicate significant difference to the control sample (Mardia Watson Wheeler tests) n.s. not significant Significance levels: ** p 

Orientation of migratory birds under ultraviolet light.

In view of the finding that cryptochrome 1a, the putative receptor molecule for the avian magnetic compass, is restricted to the ultraviolet single co...
696KB Sizes 0 Downloads 0 Views