Genetic analysis of cryptochrome in insect magnetosensitivity

Abstract

The earth's magnetic field plays an important role in the spectacular migrations and navigational abilities of many higher animals, particularly birds. However, these organisms are not amenable to genetic analysis, unlike the model fruitfly, Drosophila melanogaster, which can respond to magnetic fields under laboratory conditions. We therefore review the field of insect magnetosensitivity focusing on the role of the Cryptochromes (CRYs) that were first identified in Arabidopsis and Drosophila as key molecular components of circadian photo-entrainment pathways. Physico-chemical studies suggest that photo-activation of flavin adenine dinucleotide (FAD) bound to CRY generates a FAD^o- Trp^o+ radical pair as electrons skip along a chain of specific Trp residues and that the quantum spin chemistry of these radicals is sensitive to magnetic fields. The manipulation of CRY in several insect species has been performed using gene editing, replacement/rescue and knockdown methods. The effects of these various mutations on magnetosensitivity have revealed a number of surprises that are discussed in the light of recent developments from both in vivo and in vitro studies.

Keywords: Drosophila; circadian; cryptochrome; electron spin; migration; quantum; radical pairs.

FIGURE 1

Structure of Drosophila cryptochrome.(A) The framework for a CRY photocycle using Escherischia coli photolyase (EcPL) and Arabidopsis thaliana CRY1 (AtCRY1). Photoexcitation excites the oxidised FAD (FAD^ox) to the singlet state (¹FAD*) which then receives an electron via a chain of three Trp residues (see Figure 1B) within CRY/photolyase generating a radical pair leaving the terminal Trp minus an electron ¹ [FAD^o− TrpH^o+] in the singlet state with antiparallel electron spins which can either reverse to the ground state (FAD^ox + TrpH) or interconvert to the triplet state ³ [FAD^o− TrpH^o+] via hyperfine interactions in which the electron spins of the unpaired electrons are in parallel. In EcPL the S and T forms of this RP (RP1) can also convert to a second RP2 by deprotonation (removing H ion) of TrpH^o+ to the neutral radical, Trp^o which can return to the dark resting state (FAD^ox + TrpH) by further redox reactions (redrawn with amendments from Sheppard et al., 2017). (B) Crystal structure of dCRY regions lying close to the FAD (blue from Czarna et al., 2013, beige Zoltowski et al., 2011). The Trp triad W420, 397 and 342 are shown together with the proposed electron skipping that generates the photoreduced FAD-Trp radical pair. (C) Crystal structure of dCRY from Czarna et al (2013). The blue represent the structure from Czarna et al. compared to a previous structure in beige from Zoltowski et al (2011). There are some significant differences in the tail structure, later corrected by Zoltowski et al (2011). (B,C) reproduced with permission from Czarna et al. (2013). (D) Overview of dCRY landmarks (not to scale). PHR, Photolyase homology domain; CT, C terminal residues 490–542; CTT C terminal tail, residues 520–542; Blue stars indicate positions of Trp tetrad, Trp 342, 394, 397 and 420. The amino acid sequence of the CT is illustrated below. The calmodulin binding domain is shown in orange, residues ∼490–516 in violet. PDZ domain binding motifs in the CTT are in red, and the sole Trp in the CTT is green. The dCRYΔ transgene is missing the CRY CTT and the GFP-CRY CT transgene encodes GFP fused to residues 490–542 (see text).

FIGURE 2

Exposure to a low frequency magnetic field has a circadian phenotype in Drosophila. Top row (A–E): double plotted representative actograms from a male fly from each genotype. The left hand actogram for each genotype shows the activity of a fly exposed to a MF, whereas the right hand panel shows the activity under sham exposure. Each row of the actogram shows 2 days of activity day1 and day2, below which is represented day2 and day3, then day 3 days 4 and so on. The double headed green arrows on the left of panel A show the first 3 days in a dim blue light-dark BLD12:12 cycle, followed by 5 days in constant dim blue light, BLL (prexposure, red arrows, brown background on actogram), followed by 8 days in the same BLL conditions but with exposure (or sham) to a 3 Hz, 300 μT MF (blue arrows, yellow background). Red lines show offsets of free-running locomotor activity reflecting any change in period after exposure. (A–C) The graphs below show the corresponding mean free-running period and sems for each genotype calculated from the pre-exposure and exposure conditions in the sham and experimental groups. (A): Canton-S wild-type (B): tim > cryΔ;cry ⁰² (timGAL4; UAScryΔ;cry ⁰² ) which means CRY that is missing the CTT (last 20 amino acids) is expressed only in clock cells in a cry-null background. (C): tim > CRYCT;cry ⁰² (timGAL4; UAS-GFP-CRYCT;cry ⁰² ) which expresses only the 52 amino acid CRYCT (fused to GFP for stability) in clock neurons (tim > CRYCT;cry ⁰² ) in a cry-null background. (A,C) show significant period shortening under exposure (but not sham), whereas the MF has no significant effect on period for the cryΔ transgene (see Figure 1). (D) The cry ⁰² null mutant shows no changes in period on exposure to a MF compare to sham (upper panel). The lower panel reveals that the response of cry ⁰² flies in constant dim BL (shown as actograms in the upper panel and on a yellow background in the lower panel) is no different from the response of cry ⁰² mutants in constant darkness (gray background in lower panel). Consequently by not showing any period lengthening under BL, cry ⁰² mutants are not informative in this assay. (E) In contrast to cry ⁰² , overexpression of CRY in clock neurons (tim > cry) in a wild-type background leads to high levels of arrhythmicity in the prexposure conditions, but application of a MF significantly rescues this arrhythmicity (upper and lower panels). Furthermore, rhythmic flies on exposure to a MF have shorter periods than those rhythmic flies that are exposed to sham (not shown, see Fedele et al., 2014a). Figure 2 redrawn from Fedele et al. (2014a).

FIGURE 3

MF exposure effects on geotaxis and neuronal firing in Drosophila. (A) Results of geotaxis assay from Fedele et al. (2014b). Under 450 nm blue light wild-type Canton S (CS) flies walk upwards (negative geotaxis), but application of a MF makes them more positively geotactic generating a significantly lower climbing score as they tend to move downwards. cry ⁰² mutants show reduced negative geotaxis, so application of a MF does not change their behaviour, therefore as in circadian behaviour (Figure 2D) this result is not informative with respect to the MF. However, expressing CRY in a mutant cry ⁰² background either in all clock neurons (tim > cry) or just those that normally express CRY (cry > cry), rescues the normal negative geotactic response in sham exposed flies which is significantly reduced in MF exposed flies. Means and sems shown. Flies were exposed to a static 500 μT MF in dim blue light.*, p < 0.05, **p < 0.01, ***p < 0.001. (B,C). Results of similar experiment as in A by Bae et al (2016). Geotactic positioning (positive geotaxis) in a tube, very similar to that used by Fedele et al., 2014b (see above, A) and shown in C, at different MF intensities. The graph shows means and sems for flies exposed to the earth’s ambient field (sham, 45 μT in Korea) with increasing intensity of exposure. In this case, the results are represented as the flies’ positive geotaxis score (% flies in lower S2-S5 sections of tube shown in C, so higher scores means the flies are moving further downwards). As field intensity increases, flies move further downwards generating a higher geotactic score, as observed in Fedele et al, 2014a). ***p < 0.0005, ****p < 0.0001 compared to sham. (D) Geotaxis in a near zero field (Bae et al., 2016). Geotaxis was compared in a zero (labelled ‘b’ see C) versus ambient 45 μT (sham) field. The ambient (sham, 45 μT) field significantly enhanced positive geotaxis (open squares) compared to zero field (filled squares), so flies moved higher in zero field, in contrast to the downward (positive geotactic) effect of the ambient MF on geotaxis. ***p < 0.0005, ****p < 0.0001. Figures (B–D) reused under open access creative commons license (http://creativecommons.org/publicdomain/zero/1.0/). (E,F) Magnetic field effects on neuronal firing in Drosophila larval aCC neuron (reproduced from Figure 1 in Giachello et al., 2016 under creative common license). (E) Representative electrophysiological recordings from a larval aCC neuron expressing CRY ectopically by using the pan-neuronal elavgal4 driver driving UAS-dcry (labelled ‘CRY’) compared to control (elavGAL4 driver only). Blue light produces significant reversible depolarization in membrane potential (ΔV_m, Y-axis, black trace) that is potentiated by a 100 mT MF (red trace). The elavgal4 driver control generates an increased depolarisation to BL (black) but no enhancement by the MF (red). (F) The results from a group of larvae are shown. Means and sems **p < 0.01.

FIGURE 4

Magnetic field effects on Monarch butterfly behaviour. (A) Flight simulator with three different axes of Helmholz coils with a Monarch tethered in the middle of the arena. (B) Migrant and lab-reared Monarchs do not respond to a constant ambient MF (∼45 μT AMI, top panel) unless it is reversed (RAMI, bottom panel), when they immediately increase their wingbeats in response in full spectrum light. (C) The wingbeat response to RAMI is observed under full spectrum light but not in darkness in wild-type of DpCry1 ^−/− knockout Monarchs. Neither does DpCry1 ^−/− show the response in blue/UV light in contrast to the wild-type. Neither wild-type nor the mutant respond to RAMI under cyan/green light (480–580 nm) (Figure redrawn from Figures 1, 2 of Wan et al., 2021 under CC open access (http://creativecommons.org/publicdomain/zero/1.0/). See article for description of statistics).

FIGURE 5

Genetic analysis of magnetosensitivity in cockroaches (Bazalova et al., 2016). (A) Diagrammatic representation of the body turn phenotype in response to a rotating MF. (B) Body turns in response to a MF (∼45μT, 18 μT horizontal component) for Periplaneta americana and Blattella germanica. Sf, steady field, rf rotating field. Basic test—MF exposure, DoWr-sham double wrapped coils. Under a MF (Basic) a significant increase in turning is observed in a rotating field (rf) for both species. Y-axis represents body turns >15^o. (C). Transient double stranded (ds)RNAi knockdown of Pacry2 and BgCry2 but not Bgcry1 nor Patimeless (Patim) eliminates the MF-mediated turning response. For (B,C) **p < 0.01, ***p < 0.001. Figure redrawn and simplified from Figures 1–3 from Bazalova et al (2016) with permission.