How living systems respond to weak electromagnetic fields represents one of the major unsolved challenges in sensory biology. Recent evidence has implicated cryptochrome, an evolutionarily conserved flavoprotein receptor, in magnetic field responses of organisms ranging from plants to migratory birds. However, whether cryptochromes fulfill the criteria to function as biological magnetosensors remains to be established. Currently, theoretical predictions on the underlying mechanism of chemical magnetoreception have been supported by experimental observations that exposure to radiofrequency (RF) in the MHz range disrupt bird orientation and mammalian cellular respiration. Here we show that, in keeping with certain quantum physical hypotheses, a weak 7 MHz radiofrequency magnetic field significantly reduces the biological responsivity to blue light of the cryptochrome receptor cry1 in Arabidopsis seedlings. Using an in vivo phosphorylation assay that specifically detects activated cryptochrome, we demonstrate that RF exposure reduces conformational changes associated with biological activity. RF exposure furthermore alters cryptochrome-dependent plant growth responses and gene expression to a degree consistent with theoretical predictions. To our knowledge this represents the first demonstration of a biological receptor responding to RF exposure, providing important new implications for magnetosensing as well as possible future applications in biotechnology and medicine.
Static magnetic fields have profound and diverse effects on living organisms ranging from prokaryotes to man1,2,3,4,5,6,7,8,9. One of the best characterized involves orientation behaviour in migratory birds, which use the magnetic field for directional sensing by a process that requires light1,7. Bird magnetosensing has been proposed to occur by the so-called radical pair mechanism, whereby weak magnetic fields can alter the spin characteristics of radical pairs generated by a biological magnetoreceptor10. An evolutionarily conserved flavoprotein photoreceptor known as cryptochrome, which forms radical pairs and is localized to the bird retina1,11,12, has been proposed as such a possible magnetoreceptor.
An intriguing feature of the radical pair hypothesis in birds is the theoretical prediction that RF signals in the 1–10 MHz range should elicit the disruption of bird directional responses to the Earth’s magnetic field13,14,15. Such disruptive effects were indeed found experimentally for RF fields, remarkably even of intensities below 10 nT9 and, in the case of broad-band fields, below 1 nT16. Many of these effects can in principle be rationalized with the radical-pair mechanism14,17. Therefore since cryptochromes have been implicated in responses to static magnetic fields in organisms ranging from plants to humans, a prediction of the radical pair hypothesis is that RF magnetic fields could also affect cryptochrome responses. In this work, we explored the intriguing possibility that such fields could trigger a biological response involving Arabidopsis cryptochrome
A rapid, quantitative, and direct assay for magnetic sensitivity is the in vivo phosphorylation of Arabidopsis cryptochrome in blue light10,18. Phosphorylation results from conformational changes triggered in the receptor, whereby the cryptochrome C-terminal domain unfolds from the protein surface and becomes accessible to cellular kinases. Cryptochrome phosphorylation can be visualized on Western blots by an upward mobility shift of the phosphorylated protein. Prior studies have shown that cryptochrome phosphorylation in plant seedlings is altered as a function of the static magnetic field3,19, and is reduced at near null LLF (low -level fields). We therefore tested whether phosphorylation of the Arabidopsis cry1 receptor was also responsive to an applied weak 7 MHz RF magnetic field. This frequency was chosen as it had been previously reported to interfere both with bird navigation13,20 and with oxidative metabolic processes in mammalian cultured cells6 that have been postulated to involve flavoprotein radicals.
The experimental setup is described in detail in Methods. Briefly, a triaxial Helmholtz coil providing current along each of the three axes (x, y, z) was adjusted to set the static magnetic field parallel to the plane of growth of the seedlings, at 40 μT intensity to approximate the local geomagnetic field. Blue light LEDs were used to illuminate the sample. To generate the RF field, a single loop Helmholtz coil was placed around the sample on a rotating axis, such that an RF field could be set in a direction that was either at a parallel or a perpendicular angle to the static magnetic field (see Methods). The RF signal was 7 MHz at 2 μTrms. A low - level magnetic field (LLF or Low Level Field) of less than 200nT was generated by layering of sheets of μ-metal shielding around the sample (Methods). All experiments were performed in a dedicated darkroom with temperature at the position of the sample monitored in real time by computer throughout the course of the experiment.
Phosphorylation experiments were performed as described previously3. Four-day old dark-grown Arabidopsis seedlings on petri plates were illuminated for 90 minutes and simultaneously exposed to RF magnetic fields. These were applied either in parallel or in perpendicular to the geomagnetic field (see Methods). As the control condition, exposure was to the geomagnetic field alone (without applied RF). Finally, a series of sham experiments were conducted at each exposure condition to control for any background variation in the experimental setup (see Methods). Seedlings were then harvested and subjected to Western blot analysis with anti-Cry1 antibody to determine the cryptochrome protein upward mobility shift resulting from phosphorylation (Methods).
The results showed a significant (up to 24%) decline in response to blue light by cryptochrome in seedlings exposed to RF fields (Fig. 1). This was demonstrated by the reduced intensity of the upward-shifted, phosphorylated band in the Western blot under conditions of applied RF fields. Consistent with previous reports19, exposure to LLF conditions likewise caused a decrease in cryptochrome response (Fig. 1). Thus, an RF magnetic field has a similar effect on plant cryptochrome activation as does simply reducing the geomagnetic field to a LLF. Although our results by themselves are not proof that the same underlying mechanism is involved, it is nonetheless intriguing that an analogous effect has been documented in migratory birds13,20.
To further confirm these findings, two additional assays for cryptochrome biological activity were performed under conditions of RF exposure. These included Arabidopsis seedling hypocotyl growth inhibition and qPCR analysis of cryptochrome regulated gene expression, both of which have been reported sensitive to applied static magnetic fields3,8. For the plant growth experiments, seedling growth was monitored over 5 days under the identical illumination and magnetic field exposure conditions as for the cryptochrome phosphorylation experiments. Increased seedling growth is an indication of reduced cryptochrome biological activity, since cryptochrome mediates seedling growth inhibition under blue light3. The results of the plant growth assay showed that exposure to either parallel or perpendicular RF magnetic fields resulted in significantly increased hypocotyl length (up to 29%), and thereby reduced cryptochrome biological function, as compared to control seedlings (Fig. 2). Exposure to LLF conditions also caused increased hypocotyl length, indicating reduced cryptochrome response. Importantly, comparison of sham RF or sham LLF exposed seedlings to those at the reference geomagnetic field (GMF) condition yielded no statistical difference in phosphorylation.
Expression analysis in response to RF magnetic fields was performed on PIN1, PIN3, and AUX1 genes, which had previously been shown to be cryptochrome-regulated and responsive to static magnetic fields8. All three of these genes showed statistically significant change in expression in response to LLF exposure (Fig. 3), in this way replicating previously obtained results8. Upon exposure to RF fields, gene expression was altered to a similar degree as for LLF (Fig. 3). In summary, altered cryptochrome biological activity has been observed as a result of magnetic field exposure in three independent, unrelated biological assays; one of which (phosphorylation) provides a direct probe of the photoreceptor activation state.
Cryptochromes are proposed to be activated through flavin reduction in response to light, hereby conversion of oxidized flavin (FADox) to radical (FADH°) and reduced (FADH-) redox states triggers the conformational change leading to the biologically active form21 and ref. 22 here (see model in Fig. 4). Reduced flavin is reoxidized back to the inactive (FADox) redox state by a reaction that consumes molecular oxygen; this reaction has the potential to produce radical pair intermediates. Because only the FADH° redox state is correlated with biological activity, the response to light by cryprochrome is determined by the equilibrium concentration of FADH° under a given illumination condition. Intriguingly, prior studies have also shown that cryptochrome-dependent responsivity to magnetic fields occurs exclusively during the reoxidation phase of the photocycle3,19,23. Furthermore, a number of possible magnetically sensitive radical pairs may be formed in the course of reoxidation from fully reduced flavin (FADH-), suggesting a change in rate constant k2b as the likely magnetically sensitive step.
We therefore calculated what magnitude of change in the rate constant k2b could provide the observed magnetic field effect reported in Fig. 3. To model our experimental results in relation to the cryptochrome photocycle, we used a previously reported kinetic model which relates plant seedling growth to the average concentration of the FADH° (active) cry1 redox form24. This model uses previously deduced estimates of in vivo rate constants for cry1 flavin reduction/reoxidation steps (see Supplementary Information for details) in combination with experimental results from this study (Fig. 3). On the assumption that only k2b (formed in the course of FADH- reoxidation) is modified by the magnetic field, we calculated that a modest increase in the rate constant of approximately 20% would result in the observed biological response to LLF. This is compatible with a possible radical pair mechanism13,20.
In conclusion, this study shows that RF magnetic fields alter the biological response characteristics of the Arabidopsis cryptochrome receptor itself, similarly to the effects of a near-null magnetic field. These results are consistent with the radical pair mechanism for magnetosensing and cannot be explained by an iron – based magnetosensor13,20, although we can not exclude that unrelated magnetosensing mechanisms exist in parallel6,25,26, or that cry may not be the direct RF receptor. Since cryptochromes are found in many organisms in the biological Kingdom including in humans, this study may lead to new biomedical applications developing RF signals to elicit desired cellular responses. Our results also may have more general implications for the capacity of living organisms to respond to man-made electromagnetic noise, by analogy with broad band RF16 which has been previously shown to disrupt orientation of birds.
Materials and methods
RF magnetic field exposure setup
The setup was essentially as previously described27. The initial static magnetic field (SMF) background inside the room varied from 25 to 60 μT as measured with a gauss meter (IDR-321, Integrity Design, VT, USA)27 in all 3 axes, and therefore required tri-axial compensation to establish a uniform pre-set SMF27 in the volumes designated for culture plates within the coils27. For these experiments, two tri-axial sets of square coils were constructed in a Helmholtz configuration27. The first set allowed for the simultaneous exposure of three 100 mm petri dishes as a control to a SMF of 40 μT27. The second set served to expose cells to a SMF of 40 μT and to either parallel/perpendicular applied weak 7 MHz magnetic fields27. In both cases, the 40 μT SMF was oriented perpendicular to the plane of growth of the plants27. The experimental exposure included both groups placed within separate tri-axial coils containing a single Helmholtz loop RF loop. The RF coil was not energized for the control SMF and was energized for the RF group27.
Each square coil (45 cm each side) consisted of 20 turns of 22 AWG enamel-coated copper wire27. Each pair of square coils was axially aligned and separated by 22.5 cm in order to achieve the Helmholtz configuration27. Each pair of coils in the Helmholtz configuration was individually driven by a power supply27. Resistive circuitry was fed in a twisted pair in order to achieve the necessary compensatory SMF in the desired direction27. The SMFs were adjusted accordingly at the isocenter of each tri-axial set as measured by a gauss meter for each axis27. A 1-turn square coil (12.5 cm side) in Helmholtz configuration was built inside one of the tri-axial sets in order to superimpose magnetic fields in the RF band also with 22 AWG enamel-coated copper wires27. The geometric center of this RF coil was aligned with that of the triaxial set used for SMF compensation27. A function generator (HP33120A, Hewlett-Packard, Palo Alto, CA) established the 7 MHz magnetic signal, and the magnitude recorded in the culture-designated volume was 2±0.5 μT (RMS) after power amplification27. The RF signal was measured with a circular search probe sensor composed of 5 turns of 22 AWG enamel-coated copper wires, 1.5 cm in radius, which were connected directly to an oscilloscope via a twisted pair feeding a coaxial cable27.
The background time-varying magnetic field was measured at the center of the tri-axial sets while inside the room in the location where the experiment was to be performed with a gauss meter (IDR-210, Integrity Design, VT) in all 3 axes27. The measurements performed resembled previous observations, where the dominant spectral magnitude was recorded at 50 Hz and was below 2 μT for all cases27. The temperature was maintained at 23 °C and verified throughout the course of the experiment by computer monitoring via a themocouple placed at the position of the sample27. The environmental parameter variance was minimal during the experiments27. The dark room was utilized exclusively for these experiments and were not opened for the duration of the exposures27.
Near-null magnetic field Exposure System
The near-null magnetic field (LLF) was produced by a double layer μ-metal cylinder. The inner layer was 11.5 cm diameter, and the outer layer was 16 cm diameter, 30 cm of height. We measured the static magnetic field (SMF) at the center of cylinder, which was the position for Arabidopsis plate and the SMF intensity was lower than 200nT. The sham LLF was produced by a Helmholtz coil placed within the cylinder. Each coil consisted of 20 windings of 1 mm-diameter copper wire around a3 plastic circular frame (10 cm diameter, at a separation of 10 cm between coils) at the center3 of inner μ-metal cylinder (described in ref. 3). The current was provided to the coils to generate 40 µT static MF, which was local GMF for the sham-control experiments. The temperature was maintained at 23 °C and verified throughout the course of the experiment by computer monitoring via a themocouple placed at the position of the sample.
Plant materials and growth conditions
Arabidopsis thaliana phyA phyB mutants were used for cry1 phosphorylation and qPCR gene expression experiments, Arabidopsis thaliana transgenic over-expressing cry1 seedlings were used in hypocotyl growth tests as previously described3. Seeds were sterilized by incubation with 25% bleach for 30 min, washed 3X with sterile water, and plated on 5 mm diameter petri dishes containing 2% (W/V) sucrose, 0.5X MS salts pH 6.0 (MP Biomedicals, INC, Illkirch, France) and 0.9% (W/V) agar. Plates were maintained at 4 °C in the darkness for 48 hours, then illuminated with red light (633 nm) at 23 °C for 24 hours. For phosphorylation and qPCR assay experiments, seedlings were returned to darkness at 23 °C for 4 days. Seedlings for hypocotyl growth assays were transferred after germination to blue light test conditions for 5 days growth. Between 100 and 200 seeds per plate were used for the phosphorylation assay; 15 seeds per plate were measured for the hypocotyl growth assay. Details are as previously described in previous refs. 3,22.
Blue-light exposure system
Please see3 for complete details. Blue light was produced by LEDs with peak wavelength of 447 nm (Quadica Developments Inc., Alberta, Canada) mounted 4.5 cm above the seedlings at the center of the exposure coil (see above). The LEDs were controlled by custom built automated programmable switches (see ref. 3. for details) to provide alternating 5 min blue light / 10 min dark pulsed illumination conditions. The photon fluence of blue-light intensity for the experiment was measured by Quantum light meter (LI-185B, LI-COR, Inc., USA).
Details of this procedure are taken from3,22. The phosphorylation assay was performed as described previously3. 4-day old dark-grown Arabidopsis seedlings were exposed to treatment conditions as follows: 7 MHz RF oriented perpendicular or parallel to the Static Magnetic Field; or LLF (low level field). The control condition was the geomagnetic field at 40 μT. Illumination at all exposure conditions was identical, consisting of repeated pulses of blue-light at 60 µmolm-2s−1 for 5 min followed by 10 min darkness. This cycle was repeated 6 times for a total time of 90 min. The temperature was maintained at 23 °C and verified throughout the course of the experiment by computer monitoring via a themocouple placed at the position of the sample. Seedlings were then quick-frozen in liquid nitrogen and total protein extracted and assayed for the presence of the cryptochrome phosphorylated band by Western blotting as previously described3.
Experimental design and controls
The experimental design to demonstrate the effect of RF and LLF on cry1 phosphorylation was as previously described3. To eliminate any possibility of artifact between experiment (test) and control comparisons, the experiments were performed in duplicate sets of at least five independent trials each. The two experimental groups were designated treatment, and control group, respectively, each of which were replicated in at least five independent biological repeats.
To determine the extent of phosphorylation, the intensity of the upper, phosphorylated band from each lane of the Western blot (cry1(Pi)) (see example gel Fig. 1, upper panel) was determined using imaging software ImageJ and expressed as a percentage of the intensity of the total cry1 protein (sum of phosphorylated plus unphosphorylated cry1) in the same lane. The formula for obtaining the extent of Cry phosphorylation is thereby [cry1(Pi)]/ [cry1 (total)] x 100 yielding the percentage of phosphorylated cryptochrome per lane. Three triplicate lanes per individual experimental condition were averaged to yield the percentage of phosphorylation in one harvested sample.
For the treatment group, the mean cry1 phosphorylation value of each exposure condition (RF parallel, RF perpendicular, LLF) was compared to the mean cry1 phosphorylation in the reference static 40 μT condition designated as Geomagnetic Field (GMF). Values were expressed as the percent difference between phosphorylation in exposed and reference samples (see Fig. 1). For the control group, we compared the mean of cry1 phosphorylation in sham - exposed samples to those of cry1 phosphorylation at the reference GMF condition; (RF perpendicular sham, RF parallel sham, or LLF sham vs. GMF reference). In this control group, the experimental setup, position of samples, illumination of samples, etc. were identical to that of the test experimental group except that no RF or LLF was applied in the test conditions (ie a mock ‘test’ group).
We then performed statistical analysis to determine the percentage difference of cry1 phosphorylation within the treatment group (RF perpendicular, N = 7; RF parallel, N = 9; LLF, N = 7) and within the control group (N = 5 for sham - exposed RF perpendicular, RF parallel, and LLF) (see below).
Hypocotyl growth experiments
Seedling growth experiments were performed as in3. For each experimental determination, 15 seedlings were measured. All analysis was performed double blind, in that the person performing the measurements did not know under which condition the plates had been grown. For the treatment group, the mean value of hypocotyl length from each treatment condition (RF perpendicular, RF parallel, LLF) was compared to the mean value of hypocotyl length in seedlings exposed to the reference GMF condition. For the control group, comparisons were made between sham-treated seedlings and the reference GMF condition. Five replicate biological repeats were performed for all experiments in both the treatment and control groups (N = 5).
For statistical analysis, the significance of differential growth from each condition in the treatment group (N = 5; +RF perpendicular vs. GMF, + RF parallel vs. GMF, + LLF vs. GMF) and from each condition of the control group (N = 5; -RF perpendicular vs. GMF, -RF parallel vs. GMF, -LLF vs. GMF) was calculated.
Quantitative RT-PCR analysis of altered gene expression
QPCR analysis was performed as described8. Dark-grown etiolated Arabidopsis seedlings were plated as for the phosphorylation assay, then illuminated at 60 µmolm−2s−1 blue-light for 5 min followed by 10 min darkness. This cycle was repeated for a total time of 3 h, under the designated electromagnetic exposure conditions (RF, LLF, or the reference GMF). After quick-freezing in liquid nitrogen, total RNA was extracted using the Total RNA Miniprep Kit (New England Biolabs) and cDNA was prepared from 1 mg total RNA using SuperScript first-strand synthesis system (Thermo Fisher Scientific). Quantitative RT-PCR was performed using Luna qPCR master mix (New England Biolabs). Arabidopsis GADPH was used as the reference gene28,29. Five biological replicates were performed for each analysis (N = 5). Primers used for gene expression analysis are listed in Table 1.
Wiltschko, R. & Wiltschko, W. Sensing magnetic directions in birds: radical pair processes involving cryptochrome. Biosensors 4, 221–42 (2014).
Fedele, G., Green, E. W., Rosato, E. & Kyriacou, C. P. An electromagnetic field disrupts negative geotaxis in Drosophila via a CRY-dependent pathway. Nat Commun 5, 4391 (2014).
Pooam, M. et al. Magnetic sensitivity mediated by the Arabidopsis blue-light receptor cryptochrome occurs during flavin reoxidation in the dark. Planta 249, 319–332 (2019).
Bazalova, O. et al. Cryptochrome 2 mediates directional magnetoreception in cockroaches. Proc Natl Acad Sci USA 113, 1660–5 (2016).
Sherrard, R. M. et al. Low-intensity electromagnetic fields induce human cryptochrome to modulate intracellular reactive oxygen species. PloS Biol 16, e2006229 (2018).
Usselman, R. J. et al. The Quantum Biology of Reactive Oxygen Species Partitioning Impacts Cellular Bioenergetics. Scientific Reports 6, 38543 (2016).
Wiltschko R. Magnetic Orientation in Animals. Zoophysiology vol.33 Springer – Verlag Berlin Heidelberg 1995. Editors S.D. Bradshaw, W. Burggren, H.C. Heller, S. Ishii, H. Langer, G. Neuweiler, D.J. Randall.
Xu, C., Zhang, Y., Yu, Y., Li, Y. & Wei, S. Suppression of Arabidopsis flowering by near-null magnetic field is mediated by auxin. Bioelectromagnetics 39, 15–24 (2018).
Yoshii, T., Ahmad, M. & Helfrich-Förster, C. Cryptochrome mediates light-dependent magnetosensitivity of Drosophila’s circadian clock. PLoS Biol 4, e1000086 (2009).
Wang, Q., Barshop, W. D., Bian, M. & Vashisht, A. A. The blue light-dependent phosphorylation of the CCE domain determines the photosensitivity of Arabidopsis CRY2. Mol Plant 8, 631–643 (2015).
Chaves, I. et al. The cryptochromes: blue light photoreceptors in plants and animals. Ann Rev Plant Biol 62, 335–364 (2011).
Ozturk, N. Phylogenetic and functional classification of the Photolyase /Cryptochrome Family. J. Photochem Photobiol 93, 104–111 (2017).
Ritz, T. et al. Magnetic compass of birds is based on a molecule with optimal directional sensitivity. Biophys J. 96, 3541–3547 (2009).
Ritz, T., Thalau, P., Phillips, J., Wiltschko, R. & Wiltschko, W. Resonance effects indicate radical pair mechanism for avian magnetic compass. 2004, 429, 177-180 (2004).
Kavokin, K. et al. Magnetic orientation of garden warblers (Sylvia borin) under 1.4 MHz radiofrequency magnetic field. J.R.Soc-Interface 11, 2014051 (2014).
Engels, S. et al. Anthropogenic electromagnetic noise disrupts magnetic compass orientation in a migratory bird. Nature 509, 353–6 (2014).
Lambert, N. et al. Quantum biology. Nature Physics 9, 10–18 (2013).
Tan, S. T., Dai, C. & Liu, H. T. Arabidopsis casein kinase1 proteins CK1. 3 and CK1. 4 phosphorylate cryptochrome2 to regulate blue light signaling. Plant Cell 25, 2618–2632 (2013).
Xu, C., Lu, Y., Chen, C. & Zhang, Y. Blue light-dependent phosphorylations of cryptochromes are affected by magnetic fields in Arabidopsis. Adv Space Res 53, 1118–1124 (2014).
Hore, P. J. & Mouritsen, H. The Radical-Pair Mechanism of Magnetoreception. Annu Rev Biophys 45, 299–344 (2016).
Ahmad, M. Photocycle and signaling mechanisms of plant cryptochromes. Curr Opin Plant Biol 33, 108–115 (2016).
Hammad, M., et al. Cryptochrome mediated magnetic sensitivity in Arabidopsis occurs independently of light-induced electron transfer to the flavin Photochem & Photobiol Sci. https://doi.org/10.1039/c9pp00469f (2020).
Wiltschko, R., Ahmad, M., Nießner, C., Gehring, D. & Wiltschko, W. Light-dependent magnetoreception in birds: the crucial step occurs in the dark. J R Soc Interface. 13, 118 (2016).
Procopio, M. et al. Kinetic Modeling of the Arabidopsis Cryptochrome Photocycle: FADH(o) Accumulation Correlates with Biological Activity. Front Plant Sci. 7, 888 (2016).
Dhimana, S. K. & Galland, P. Effects of weak static magnetic fields on the gene expression of seedlings of Arabidopsis thaliana. Journal of Plant Physiology 231, 9–18 (2018).
Qin, S. et al. A magnetic protein biocompass. Nature Materials 15, 217–226 (2016).
Usselman, R., Hill, I., Singel, D. & Martino, C. (2014). Spin Biochemistry Modulates Reactive Oxygen Species (ROS) Production by Radio Frequency Magnetic Fields. PLoS ONE 9(3), e93065 (2014).
Liu, H. et al. The RING-Type E3 Ligase XBAT35.2 Is Involved in Cell Death Induction and Pathogen Response. Plant Physiol 175, 1469–1483 (2017).
Ritz, T., Adem, S. & Schulten, K. A Model for Photoreceptor-Based Magnetoreception in Birds. Biophys. J. 78, 707–718 (2000).
Agliassa, C., Narayana, R., Christie, J. M. & Maffei, M. E. Geomagnetic field impacts on cryptochrome and phytochrome signaling. J Photochem Photobiol B 185, 32–40 (2018).
We are indebted to Brendon Green for help with figures, and to Adam Purvis, Dorothy Engle, Justin Link and Haider Raad for valuable assistance. Support was from NSF (#1658640), AFOSR (FA9550-14-0-0409) and AFOSR (FA9550-17-1-0458).
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Albaqami, M., Hammad, M., Pooam, M. et al. Arabidopsis cryptochrome is responsive to Radiofrequency (RF) electromagnetic fields. Sci Rep 10, 11260 (2020). https://doi.org/10.1038/s41598-020-67165-5
Scientific Reports (2021)
Journal of Plant Biochemistry and Biotechnology (2020)