Keywords

20.1 Introduction

The use of directional information from the Earth’s magnetic field for compass orientation has been demonstrated in a large variety of animals across different taxa. Behavioral and physiological evidence suggests two possible mechanisms that could independently allow the detection of magnetic properties (for reviews see Wiltschko and Wiltschko 1995, 2005; Freake et al. 2006; Johnsen and Lohmann 2008): (1) a light-dependent biochemical process detecting the axial alignment and inclination angle of the geomagnetic field lines, providing the animals with directional magnetic information for magnetic compass orientation (inclination compass), and (2) a magnetite-mediated process, providing magnetic map information from spatial variation in the intensity and/or inclination of the geomagnetic field (map or signpost sense). The functional properties of the magnetic compass of most animals studied to date can be assigned to either one of the two groups, consistent with either a light-dependent mechanism or a non-light-dependent, magnetite-based mechanism. This chapter focuses on the light-dependent process of magnetoreception, summarizing the state of the art of behavioral, physiological, neurobiological, and biophysical evidence supporting the involvement of light in magnetic compass orientation.

20.2 Behavioral Evidence for Light-Dependent Magnetoreception

Behavioral experiments have conclusively identified light to be a key prerequisite for the primary magnetoreception process in a growing number of organisms including birds, amphibians, and some insects (Phillips and Borland 1992a; Phillips and Sayeed 1993; Wiltschko et al. 1993, 2010; Deutschlander et al. 1999b; Muheim et al. 2002; Vacha and Soukopova 2004; Phillips et al. 2010a). In addition, there is strong (albeit indirect) evidence that also C57BL/6J mice have a light-dependent magnetic compass (Muheim et al. 2006). Magnetic compass orientation independent of light has only been shown in a few exceptions, like the subterranean mole rats (Marhold et al. 1997) and sea turtles (Lohmann and Lohmann 1993). In the case of the mole rats, magnetic compass information has been shown to be mediated by a magnetite-based mechanism (Thalau et al. 2006). It is well possible that these exceptional cases are the result of an adaptation to the specific lifestyles of these organisms. As a general conclusion drawn from recent knowledge, animals need access to light of a specific wavelength range and intensities to be able to perceive magnetic compass information.

20.2.1 Light-Dependent Magnetic Compass Orientation in Insects

Magnetic compass orientation in adult and larval Drosophila and mealworm beetles (Tenebrio spp.) depends on the wavelength of light: animals trained towards a light gradient under UV light orient towards the trained magnetic direction of the light gradient when tested in a uniform arena under short-wavelength light (<450 nm). When tested under long-wavelength light (>450 nm), they shift their orientation by ~90° relative to the trained magnetic direction (Phillips and Sayeed 1993; Dommer et al. 2008; Vacha et al. 2008). This suggests that these insects possess a light-dependent magnetic compass with antagonistic short- and long-wavelength inputs. The spectral dependencies of magnetic compass orientation in both Drosophila and Tenebrio are compatible with a cryptochrome-based mechanism in which the magnetic field has antagonistic effects on short- and long-wavelength photo-signaling pathways (see below; reviewed by Phillips et al. 2010a).

20.2.2 Antagonistic Spectral Mechanism Mediating Magnetic Compass Orientation in Amphibians

Similar 90° shifts in orientation under longer wavelengths have also been described in newts and frog and toad tadpoles (Phillips and Borland 1992b; Freake and Phillips 2005; Diego-Rasilla et al. 2010, 2013; reviewed by Phillips et al. 2001, 2010a). The shoreward orientation of North American Eastern red-spotted newts trained to learn the magnetic alignment of an artificial shore is mediated by a magnetic inclination compass that depends on the wavelength of the ambient light. Magnetic orientation of newts tested under long-wavelength light (>500 nm) is rotated by 90° from the trained shoreward directions shown under short-wavelength light (400–450 nm; Phillips and Borland 1992a, b). Newts tested under an intermediate wavelength of 475 nm are disoriented. It is highly unlikely that this abrupt shift from oriented behavior to complete disorientation is caused by a gradual decrease in sensitivity of one spectral mechanism alone. This led to the suggestion that two antagonistic spectral mechanisms might be involved (Phillips and Borland 1992b; Deutschlander et al. 1999b): one high-sensitive, short-wavelength mechanism mediating oriented behavior in the trained direction of the artificial shore (Fig. 20.1a) and a second low-irradiance, long-wavelength mechanism shifting orientation by 90° relative to the shore (Fig. 20.1b). In the intermediate-wavelength condition, both mechanisms are equally excited, which leads to disorientation (Fig. 20.1c). Complete elimination of magnetic compass orientation in newts tested under near-infrared light (>715 nm) adds additional support for the hypothetical involvement of two discrete spectral mechanisms (Phillips and Borland 1992c).

Fig. 20.1
figure 1

Illustration of the hypothetical antagonistic spectral mechanism proposed to mediate the light-dependent array of axially sensitive receptors. (a–c) Magnetic modulation patterns as suggested to be perceived by an animal looking parallel along the magnetic field vector (for more details, see Fig. 20.4). (d–f) Alignment of putative receptors (small rectangles) along the magnetic axis in a spherical receptor organ, like the eyes or pineal organ. (g) Putative spectral sensitivity curves for two hypothetical antagonistic inputs to the light-dependent magnetic compass with a short- and a long-wavelength mechanism. Under short-wavelength light (a–d), the receptors aligned along the magnetic axis show a decrease in response (dark rectangles labeled “−”), while receptors aligned perpendicular to the magnetic field (gray rectangles) remain unaffected. Under long-wavelength light (c–e), the receptors aligned along the magnetic axis show an increase in response (white rectangles labeled “+”), resulting in a “reversed” magnetic modulation pattern. Arrowheads at the edge of the circular array indicate the axes with the lowest level of response, which differ by 90° in (a–d) and (c–e). (b–d) Under intermediate wavelengths of light, both mechanisms are activated equally and the directional response is eliminated (Adapted from Phillips et al. (2010a))

Full size image

20.2.3 Magnetic Compass Orientation of Birds Depends on the Presence, Wavelength, and Irradiance of Light

First evidence for the importance of light for magnetic compass orientation in birds came from homing experiments with pigeons. Young, inexperienced homing pigeons were disoriented after displacement in complete darkness or under 660 nm red light but oriented towards the home direction when transported to the release site under 565 nm green or full-spectrum light (Wiltschko and Wiltschko 1981, 1998). Orientation experiments with passerine migrants in orientation funnels illuminated with monochromatic light gave further support for the involvement of a wavelength-dependent, light-sensitive magnetoreception process. Night-migratory songbirds, like European robins (Erithacus rubecula), Australian silvereyes (Zosterops lateralis), and garden warblers (Sylvia borin), showed well-oriented behavior into the seasonally expected migratory directions when tested under monochromatic light of peak wavelengths between 373 nm (UV) and 565 nm (green), comparable to the responses shown under full-spectrum light (Rappl et al. 2000; Wiltschko and Wiltschko 2001; Wiltschko et al. 2010; Fig. 20.2). However, when tested under longer-wavelength spectra, the birds became disoriented (Wiltschko et al. 1993; Wiltschko and Wiltschko 1999; Fig. 20.2). These results indicate that light-dependent magnetoreception becomes critical under light of peak wavelengths longer than 565 nm.

Fig. 20.2
figure 2

Spectral curves under which migratory songbirds have been tested for magnetic compass orientation. The birds are well oriented when tested under low irradiance (0.5–21 × 1015 quanta s−1 m−2) to green light (solid spectral curves) but become disoriented under longer-wavelength light (dotted spectral curves)

Full size image

In a series of experiments with European robins tested under narrowband (9–11 nm half bandwidth) green (560.5 nm) and green-yellow (567.5 nm) light, Muheim et al. (2002) showed that the transition from oriented magnetic compass behavior under 560.5 nm to disorientation under 567.5 nm happens very abruptly (Fig. 20.3). The authors concluded that at least two spectral mechanisms are underlying light-dependent magnetic compass orientation in birds (Muheim et al. 2002) and proposed an antagonistic mechanism similar to that found in amphibians and insects.

Fig. 20.3
figure 3

Magnetic orientation of European robins tested during autumn migration under narrowbanded light. Birds are well oriented under 560.5 nm green light (left) and totally disoriented under 567.5 nm green-yellow light (middle) and showed a 90° shift in orientation under 617 nm red light (right) (Adapted from Muheim et al. (2002))

Full size image

Light-dependent magnetoreception varies in birds not only with wavelength but also with photon irradiance (light intensity) at the same wavelength. Night-migratory birds tested under low-light regimes (~0.5–21 × 1015 quanta s−1 m−2), similar to light levels experienced under a starry sky, are well oriented towards the migratory direction. Birds tested under higher photon irradiances (~29–54 × 1015 quanta s−1 m−2) are disoriented or show shifts in orientation deviating from the expected migratory direction (e.g., Wiltschko and Wiltschko (2001); Muheim et al. (2002); Wiltschko et al. (2010)). At such irradiances, birds do not seem to be able to use their inclination compass, that is, they do not react to an inversion of the vertical component of the magnetic field (Wiltschko et al. 2003a) and their magnetic compass orientation performance cannot be disrupted by radio-frequency electromagnetic fields (Wiltschko et al. 2005). Instead, the birds show a reaction to shifts of the horizontal component of the magnetic field vector, indicating the use of a polarity compass or an alignment along a magnetic direction.

There is general consensus on the involvement of at least two magnetoreception mechanisms, which are used under different wavelength scenarios. However, there is a vivid debate over the underlying mechanisms and putative interaction of these receptor types. Based on the parallels between the findings in behavioral experiments on birds, insects, and newts, Muheim and coworkers suggested the presence of two antagonistically interacting spectral mechanisms, with a high-sensitive short-wavelength mechanism leading to oriented behavior and a low-sensitive long-wavelength mechanism leading to shifted orientation (Muheim et al. 2002). Similar to the behavior observed in newts, the disorientation found in migratory birds under the intermediate wavelengths could be explained by an equal excitation of both magnetoreceptor types (Phillips et al. 2010a). Alternatively, over-excitation of both mechanisms could make the perceived magnetic pattern unrecognizable; thus, the birds would not be able to use the vision-mediated inclination compass any longer and would need to resort to a light-independent magnetite-based receptor. However, this hypothesis is opposed by the view that the observed 90° shifts under red are no true 90° shifts, but instead “fixed” orientations (absolute geographic 90° shifts during autumn and spring both turned to the same geographic direction and thus not relative to the natural migratory direction). This would reject the assumed similarity of the magnetoreception system in birds and newts (Wiltschko et al. 2004a). Instead, it has been suggested that the shifted orientation responses observed under high light irradiances are the results of a third, yet unspecified, novel type of light-dependent magnetoreceptor. This receptor would not be based on a light-mediated radical-pair mechanism nor a magnetite-based process, since the “fixed” orientation responses under such conditions differ between wavelengths and are not mediated by the inclination compass any longer (Wiltschko et al. 2005).

20.3 Mechanisms of Light-Dependent Magnetoreception

The behavioral data presented above imply that the light-mediated magnetic information necessary for compass orientation has to meet the following requirements: (1) The nature of magnetic information used for compass orientation has to be axial rather than polar, that is, sensing the alignment of the magnetic field lines in space but not the polarity of flux. (2) The functional range of the magnetic compass has to be reasonably plastic in order to adapt to different magnetic field intensities outside to locally experienced intensity range. This was shown in European robins that were disoriented when tested in artificial magnetic fields with intensities weaker (4–34 μT) or stronger (60–105 μT) than the local magnetic field at the testing locality (47 μT; Wiltschko 1968, 1978; Wiltschko et al. 2006; Winklhofer et al. 2013). However, after preexposure to such unnatural, and never before experienced, magnetic fields for just a few hours, the very same birds were able to orient using their magnetic compass perfectly fine in subsequent experiments. Birds thus seem to be able to learn or adapt to changing properties of the ambient magnetic field, that is, they can learn to orient under novel magnetic conditions. In other words, the functional range of their magnetic compass is flexible and allows adjustment to previously not experienced magnetic conditions, allowing magnetic compass orientation as long as the magnetic field provides directional information.

20.3.1 Chemical Magnetoreception Based on a Radical-Pair Mechanism

Various biophysical mechanisms have been suggested to explain light-dependent magnetoreception in animals. The currently most discussed model was first proposed by a physicist, Klaus Schulten. He suggested that the yield of a biochemical reaction proceeding via a radical pair might be dependent to the orientation of an external magnetic field and could thus be used for compass orientation (Schulten et al. 1978; Schulten 1982). The proposed mechanism involves a light-induced electron transfer between an electron donor and an acceptor molecule. The electron transfer results in the generation of a transient radical-pair intermediate that can exist in a singlet or a triplet excited state, and the transient radicals will subsequently decay in chemically different singlet or triplet products. Theoretical calculations and in vitro experiments have shown that the ratio between singlet and triplet products from radical-pair reactions is sensitive to and can thus be modulated by an Earth-strength magnetic field, thereby theoretically providing the basis for a magnetic compass (Henbest et al. 2008; Maeda et al. 2008, 2012; Rodgers and Hore 2009). This idea was revived in 2000 by Klaus Schulten together with Thorsten Ritz. The physicists proposed that the retina with its almost perfect half-ball shape would be well suited as an ordered structure and that the radical-pair intermediate could be involved in the visual reception system, for example, by altering the photosensitivity of the photoreceptor molecule (Ritz et al. 2000). Ritz further proposed cryptochromes as likely photoreceptor molecules. According to this hypothesis, the reaction yield anisotropy of the receptor radical pair governs the directional response. Magnetosensitive photoreceptors arranged in an ordered array in photosensory organs, such as the retina or the pineal, would respond differently, depending on their alignment relative to the magnetic field. Consequently, the effect on the visual perception process would be dependent on the alignment of the photoreceptor to the geomagnetic field and thus allow the animals to derive a directional heading by literally “seeing” the magnetic field lines (Fig. 20.4). In such a light-sensitive magnetoreception system, the animals could perceive the magnetic field as a three-dimensional pattern of light irradiance or color variation in their visual field (Fig. 20.4; for recent reviews, see Rodgers and Hore 2009; Ritz et al. 2010; Solov’yov et al. 2010; Phillips et al. 2010b).

Fig. 20.4
figure 4

Illustration of hypothetical light-dependent magnetic compass perception through magnetosensitive photoreceptors. (a left) Three-dimensional pattern of the magnetic field vector consisting of a dark area on each side of the magnetic field axis and a ring in the center. (a right) Magnetic modulation pattern perceived by an animal (bird) looking parallel along the magnetic field vector (arrow). (b) Magnetic modulation patterns at different latitudes (i.e., different magnetic field inclinations) in the Northern and Southern Hemisphere. (c) Magnetic modulation patterns perceived by animals facing different directions relative to the alignment of the magnetic field during a head scan from west to east (After Ritz et al. 2000)

Full size image

The general feasibility of radical-pair-based magnetoreception has recently been demonstrated by spectroscopic observations of an artificially designed molecule system as model magnetoreceptor: these data clearly show that the radical-pair reaction process is sensitive to geomagnetic field strengths and the lifetime of a light-induced transient radical pair is long enough to be altered by Earth-strength magnetic fields. The model receptor molecule provides further insight into the structural and dynamic design that is required to detect directional information of the Earth’s magnetic field necessary to facilitate magnetic compass orientation (Maeda et al. 2008).

20.3.2 Behavioral Evidence for a Radical-Pair-Based Magnetoreceptor

The radical-pair model meets all the necessary criteria to be a likely candidate for the light-dependent magnetic compass receptor (Ritz et al. 2002, 2010; Liedvogel and Mouritsen 2010; Phillips et al. 2010a; Mouritsen and Hore 2012): (1) the putative visual image perceived by the animal has axial properties, thus does not allow determination of the polarity of the field lines and is in line with an inclination compass and (2) the singlet-triplet yield and, consequently, the visual pattern perceived by the animal depends on the intensity of the magnetic field; thus, exposure to magnetic field intensities never experienced before can lead to disorientation, followed by a slow adaptation to the “new” pattern (cf. Wiltschko 1968, 1978; Wiltschko et al. 2006; Winklhofer et al. 2013). Adaptation has also been suggested as explanation for the ability of birds to orient under monochromatic red light after preexposure to the same light for one hour before the start of the orientation experiments (Wiltschko et al. 2004b).

The most convincing evidence for the involvement of a radical-pair mechanism in light-dependent magnetoreception comes from experiment using low-intensity oscillating radio-frequency magnetic fields (RF fields) in the lower MHz range (0.1–100 MHz). This treatment serves as a diagnostic tool that allows testing whether a radical-pair mechanism is involved in the primary magnetoreception process of an orientation response (Henbest et al. 2004; Ritz et al. 2004). The reasoning for this is that high-frequency oscillatory fields should be oscillating too fast and be too low in intensity to affect a magnetite crystal receptor (Steiner and Ulbricht 1989; Kirschvink 1996). However, they should interfere with the electrons, which are involved in the radical-pair forming process. These properties make the application of RF fields a unique tool to study the involvement of a radical-pair mechanism in magnetoreception. Thus, if RF disturbance leads to a behavioral response – either resulting in disorientation or a directional shift in magnetic orientation behavior – this treatment proves that whatever is involved in the process of magnetic orientation does involve a radical-pair forming process. In contrast, if magnetic compass orientation is solely based on a magnetite-based process, magnetic compass orientation should remain unchanged by the RF treatment.

RF interference has been applied to test the magnetic compass orientation performance of birds and mole rats: European robins and zebra finches exposed to RF fields aligned nonparallel to the geomagnetic field vector became disoriented when tested under either a broadband RF field or distinct single frequencies of 1.375 or 7 MHz (Ritz et al. 2004; Thalau et al. 2005; Keary et al. 2009). On the other hand, the orientation of mole rats, which is supposed to be mediated by a non-light-dependent magnetoreceptor, was not impaired by a RF field, supporting previous indications that their magnetic compass is light independent and based on magnetite (Thalau et al. 2006). These data provide the so far strongest, albeit indirect, support of a radical-pair-based magnetoreceptor mechanism in birds.

20.3.3 Involvement of Cryptochromes in Magnetosensitive Photoreception

A photoreceptor molecule involved in the primary magnetoreception process mediated by a light-dependent radical-pair mechanism as described above needs to meet a number of criteria. The putative receptor molecule needs to form radical pairs that persist long enough so that the radical-pair yields can be modified by an Earth-strength magnetic field and localization in a spatially fixed relationship relative to each other (reviewed by Rodgers and Hore 2009; Liedvogel and Mouritsen 2010; Phillips et al. 2010a; Mouritsen and Hore 2012). The classical photopigments, like the opsins, do not possess the biochemical properties to form radical pairs and thus are unlikely to be involved in magnetoreception. In contrast, there are strong indications that cryptochromes are the thought-after candidates for the magnetoreception molecule. Cryptochromes are blue-green light photopigments, mainly known for their role in the animal circadian clock (e.g., Emery et al. 1998; Cashmore et al. 1999; Partch and Sancar 2005; Chaves et al. 2011). They are the only currently known vertebrate photoreceptor molecules that have the potential to form radical pairs upon excitation by light. The ancestral forms of cryptochromes, the photolyases, have been demonstrated to form radical-pair intermediates that persist long enough for magnetic field effects to occur (Weber et al. 2002; Giovani et al. 2003). More recently, also cryptochromes have been shown to produce persistent, spin-correlated radical pairs upon photoexcitation in several taxa of the animal kingdom, including insects, amphibians, and birds (Liedvogel et al. 2007b; Biskup et al. 2009; Schleicher et al. 2009).

Cryptochromes are ubiquitous and have been demonstrated in a large variety of organisms, including bacteria, plants, and animals, both invertebrates and vertebrates, including humans (for a review, see Chaves et al. 2011). In species known to use a magnetic compass, cryptochromes have been reported in Drosophila melanogaster (Emery et al. 1998), bullfrogs (Eun and Kang 2003), laboratory mice (Van der Horst et al. 1999), mole rats (Avivi et al. 2004), and birds. To date, cryptochrome expression has been reported in the retinae of two migratory bird species, European robins and garden warblers, and several nonmigratory species, including chicken (Gallus gallus) and zebra finches (Taeniopygia guttatta; Möller et al. 2004; Mouritsen et al. 2004; Niessner et al. 2011; reviewed by Liedvogel and Mouritsen 2010). A detailed neurohistological study of cryptochrome expression in robins and chicken recently revealed that the outer segments of the avian UV/V cones are most likely the primary sites of the light-dependent magnetoreceptor (Niessner et al. 2011). Niessner and colleagues found cryptochrome expression in virtually every cone across the entire retina, which is one of the requirements for the radical-pair model to work.

Recent studies with Drosophila melanogaster convincingly showed that cryptochromes are involved in light-dependent magnetoreception. Gegear and coworkers found that adult Drosophila are able to discriminate a magnetic field that is about ten times the intensity of the Earth’s magnetic field (Gegear et al. 2008). This ability requires both a functional cryptochrome gene and broadband illumination that includes short-wavelength (λ < 420 nm) light. Also, cryptochrome-mediated effects of blue light on the circadian rhythm in Drosophila were shown to be influenced by magnetic fields (Yoshii et al. 2009). It has even been shown that the human variant of the cryptochrome can rescue light-dependent magnetosensitivity in a Drosophila mutant lacking Drosophila cryptochrome (Foley et al. 2011). Together, these findings suggest that magnetic field detection may be an intrinsic property of cryptochrome-based photo-signaling systems, from which light-dependent magnetoreception might have evolved (Phillips et al. 2010a). Still, more research – especially within naturally occurring field strengths – is needed to unambiguously separate observations of cryptochrome expression as a result of circadian rhythmicity and magnetoreception.

20.3.4 Is There a Common, Cryptochrome-Based Mechanism Underlying Light-Dependent Magnetoreception in Insects, Amphibians, and Birds?

Magnetic compass orientation in newts shows an abrupt transition from normal to 90° shifted orientation at about 475 nm, which has been suggested to be mediated by an antagonistic interaction of short- and long-wavelength inputs (Phillips and Borland 1992b; Deutschlander et al. 1999a; Phillips et al. 2001). In aquatic vertebrates, <480 nm light converts the fully oxidized form of the cryptochrome flavin chromophore (FADox) into the radical, flavo-semiquinone form (FADH•), which in turn is excited by light between 480 and 650 nm (Zikihara et al. 2008; Biskup et al. 2009; Fig. 20.5a–c). This nicely corresponds with the spectral properties of magnetic compass orientation in newts, which made Phillips and coworkers propose that the fully oxidized and flavo-semiquinone forms of cryptochrome could provide the short- and long-wavelength inputs to the light-dependent magnetic compass, that is, providing antagonistic magnetically sensitive inputs (Phillips et al. 2010a). In view of the observation that amphibian cryptochromes can form spin-correlated radical pairs (Biskup et al. 2009), it makes cryptochrome molecules suitable candidates for an involvement in a radical-pair mechanism.

Fig. 20.5
figure 5

(a) Interconversion of the three redox states of the flavin chromophore in a typical cryptochrome photosystem. The fully oxidized flavin chromophore (FADox) is reduced by ~370–525 nm light to the partially reduced flavo-semiquinone radical (FAD•− or FADH•) via a transient FAD*. Upon illumination by ~400–610 nm light, the flavo-semiquinone radicals can either convert to the fully reduced form (FADH- or FADH2) or alternatively be reverted back to the fully oxidized form (FADox). A magnetic field effect on photoreduction of the fully oxidized flavin chromophore (FADox) has been generally accepted to be involved in the light-dependent magnetic compass. The photoreduction of the flavo-semiquinone radical (FAD•− or FADH•) to the fully reduced form has been proposed to be involved in the antagonistic effect of the magnetic field on the response of the magnetic compass under long-wavelength light in amphibians and some insects (Phillips et al. 2010a). (b–d) Absorption of (b) insect cryptochrome (Drosophila DmCRY1; Berndt et al. 2007), (c) zebra fish cryptochrome (ZfCRY-DASH; Zikihara et al. 2008), and (d) chicken cryptochrome (GgCRY4; Ozturk et al. 2009). The solid lines give the absorption of the fully oxidized state (FADox) and the dotted lines the absorption of the partially reduced form of the flavo-semiquinone radical (FAD•− or FADH•). The gray highlighted wavelength ranges indicate changes in magnetic orientation behavior found in insects (b), amphibians (c), and birds (d) (Simplified after Phillips et al. (2010a))

Full size image

The behavioral and neurophysiological responses of insects to magnetic stimuli under UV and visible light are also consistent with the action spectra of photo-signaling pathways involving different redox forms of insect cryptochromes (Phillips et al. 2010a; Fig. 20.5b). Here, FAD is reduced to the anion radical FAD•− upon blue-light excitation, which has been suggested to be the ground state (Ozturk et al. 2009). Exposure to darkness leads to a complete reoxidation. In addition, a spectral antagonism has been shown to be involved in cryptochrome photo-signaling systems in vivo in both plants (Folta and Maruhnich 2007) and Drosophila (Hoang et al. 2008). In these systems, the short-wavelength-dependent responses are presumed to reflect a signaling pathway involving photoreduction of the fully oxidized to the radical form, while the antagonistic effects of long-wavelength light is suggested to further photoreduce the radical form to the fully reduced form, which is subsequently converted back into the fully oxidized form in the dark (Berndt et al. 2007; Hoang et al. 2008).

Still, it remains to be shown whether the long-wavelength-dependent photoreduction of the cryptochrome radical to the fully reduced form would be antagonistic to its effect on the initial photoreduction reaction, that is, whether the magnetic field would produce inverse or complementary patterns of response (Phillips et al. 2010a; Fig. 20.5).

In birds, an involvement of cryptochrome is less well clear. The well-oriented behavior under UV to 565 nm green light, abrupt transition to disorientation under 568 nm green-yellow light, and 90° shifted orientation or disorientation under longer wavelengths (Muheim et al. 2002; Wiltschko et al. 2010) fit the cry absorption spectrum only to a limited degree (Figs. 20.5d). Most cryptochromes in their fully oxidized form are insensitive above 500 nm (Liedvogel et al. 2007b), while birds are well oriented towards the migratory direction also under 565 nm green light (Muheim et al. 2002; Wiltschko et al. 2010). Also the change in orientation shown by birds around 565–570 nm is not reflected by the transition between neither of the known redox forms of bird cryptochromes (Liedvogel et al. 2007b; Ozturk et al. 2009). Thus, it remains unclear how the magnetic compass responses of birds under different light spectra can be explained by a cryptochrome-based magnetic compass.

20.4 Localization of the Light-Dependent Magnetoreceptor

Where in the animals are the light-dependent magnetoreceptors located, by which neuronal pathway is the information transmitted from the receptor to the brain, and where in the brain is the information processed? Magnetic fields can transmit through any type of organic matter; thus, magnetoreceptors can be located just about anywhere in animals’ bodies. However, as outlined above, behavioral experiments clearly show that light is necessary for light-dependent magnetic orientation to function; thus, the receptors have to be located at a peripheral site of the animals that can be reached by light.

20.4.1 Extraretinal Receptor Sites for Light-Dependent Magnetoreceptor in Amphibians

In animals that have special photoreceptors-containing structures apart from the eyes, like the parietal eyes of reptiles or the pineal complex of fish, some reptiles, and the frontal organ of amphibians, light-dependent photoreception may take place in extraretinal photoreceptors (for review, see Phillips et al. 2010a). This has been elegantly demonstrated in Eastern red-spotted newts. Light-dependent magnetoreception of the shoreward orientation of newts takes place in photosensitive, extraocular photoreceptors in the pineal complex or hypothalamus (Deutschlander et al. 1999a). Newts trained to learn the magnetic alignment of the artificial shore in the training tank with the top of their head, but not their eyes, covered with a red-light filter demonstrate the same 90° shifted response shown by newts tested completely illuminated by red light (Deutschlander et al. 1999a). Thus, the light-sensitive-magnetosensitive receptors mediating shoreward orientation in newts are most likely located in extraocular photoreceptors in the pineal complex or hypothalamus. In anuran amphibians and lizards, photoreceptors with two antagonistic photoreception mechanisms, like those proposed to underlie the light-dependent magnetic compass in newts, have been found in the pineal complex (Eldred and Nolte 1978; Solessio and Engbretson 1993).

20.4.2 The Eyes as Sites for Light-Dependent Magnetoreception in Birds and Mammals

In birds, and also mammals, the only locations where light can reach specialized photoreceptors are the eyes. Unlike in newts, an involvement of the avian pineal in magnetoreception is very unlikely, despite of responsive cells in the avian pineal to magnetic field inversions (Semm 1983; Demaine and Semm 1985). Homing pigeons homed successfully after pinealectomy, while pinealectomized pied flycatchers (Ficedula hypoleuca) showed seasonally appropriate orientation as long as they received daily injections of melatonin, suggesting no crucially dependent involvement of the pineal in magnetoreception (Maffei et al. 1983; Schneider et al. 1994).

Extracellular electrophysiological recordings provided first evidence for the involvement of the visual center in light-dependent magnetoreception. Cells in the nucleus of the basal optic root (nBOR) and in the optic tectum showed magnetic responsiveness to changes in the direction of a magnetic field and to slow inversions of the vertical component of the magnetic field, with peak responses under wavelengths of 503 nm and 582 nm (Semm et al. 1984; Semm and Demaine 1986). The nBOR receives direct input from the contralateral retina, thus supporting the hypothesis that light-dependent magnetoreception takes place at locations innervated by the optic nerve, with the eyes as likely candidates. Unfortunately, these studies have been proven difficult to replicate; thus, any conclusions drawn from these findings have to be taken with caution.

Still, recent research on the neural basis of magnetoreception has largely substantiated these findings and provided new insights into the neural pathways and brain areas involved in information transfer and processing in both birds and mammals (e.g., Nemec et al. 2001, 2005; Heyers et al. 2007). A brain structure in the visual Wulst, named “Cluster N,” connected to the retina via the thalamofugal pathway, has recently been identified and suggested to be involved in the processing of light-dependent magnetic information in migratory birds during nighttime (Mouritsen et al. 2005; Heyers et al. 2007; Zapka et al. 2010). Migratory birds with a (chemically) lesioned Cluster N were shown to be disoriented when tested for magnetic compass orientation, while their sunset and star compass remained intact and functional for orientation. These results strongly indicate that Cluster N is involved in processing magnetic compass information at nighttime under low light levels (Zapka et al. 2009).

As there is almost complete crossover of the fibers of the optic nerves, the left brain hemisphere gets its visual input almost exclusively from the right eye and vice versa. To test whether magnetic compass orientation is also lateralized, either one of the experimental birds’ eyes is unilaterally covered with light-tight eye cops prohibiting any stimulus being perceived via this respective eye, and the birds are then tested for magnetic compass orientation. Such lateralization experiments initially suggested that magnetic compass information is laterally processed with a dominance of the right eye and left brain hemisphere (Wiltschko et al. 2002, 2003b). However, follow-up experiments could not replicate these findings, neither on the receptor level nor on the level of higher integration areas in the brain, but rather suggest that both garden warblers and European robins have a magnetic compass in both eyes (Liedvogel et al. 2007a; Hein et al. 2010; Engels et al. 2012).

20.5 Outlook

During the past decade, new techniques and highly interdisciplinary research approaches have led to significant advances in our understanding of the biophysical and molecular mechanisms of light-dependent magnetoreception. The use of inducible transcription factors to study neuronal activity during magnetoreception, the search for cryptochrome expression at potential sites of magnetoreception, and the study of orientation behavior under RF fields all have provided valuable, but nevertheless indirect, evidence for the location, type, and biochemical and molecular mechanism of the magnetoreceptor.

With the development of reliable magnetic compass assays in model organisms such as Drosophila melanogaster (Phillips and Sayeed 1993; Dommer et al. 2008; Painter et al. 2013), zebra finches (Voss et al. 2007; Keary et al. 2009), and laboratory mice (Muheim et al. 2006), access to cryptochrome-deficient model animals has opened up another new and promising avenue in magnetoreception research, allowing direct tests of the involvement of cryptochromes or other molecules involved in the primary magnetoreception process.

To understand the type of magnetic stimulus and exact functionality of candidate brain regions that are involved in magnetic orientation, carefully controlled electrophysiological experiments are needed. Electrophysiology is extremely hard to carry out (and prone to artifacts) with magnetic stimuli involved, and all data reported to date have been proven difficult to repeat. However, these approaches will be key to (1) definitively prove the involvement of candidate brain areas in integrating magnetic information, (2) identify and characterize the exact type of (magnetic) stimulus that these brain areas most sensitively respond to (it could be a sudden change or a gradual variation of the azimuth, that is, the horizontal direction, of the resulting magnetic vector, or it could be an inversion of the inclination angle of the magnetic field vector), (3) understand the nature of (inhibitory/excitatory) neuronal responses (so far, nothing is known about the intrinsic neuronal dynamics of, for example, Cluster N), and (4) provide insight into the temporal scale of the putative response (pattern) after stimulus onset.