The compass within the clock – Part 2: Does cryptochrome radical-pair based signalling contribute to the temperature-robustness of circadian systems?


 For nearly fifty years, various experiments have repeatedly established the influence of the geomagnetic field (GMF) on the circadian system of various organisms, including humans. Such investigations of circadian-GMF interactions have been primarily evaluated under the hypothesis that the GMF may be acting as a “secondary zeitgeber” for circadian systems. However, there are logical and experimental grounds to question such notions – these are outlined in an initial paper “The compass within the clock – Part 1: The hypothesis of magnetic fields as secondary zeitgebers to the circadian system – logical and scientific objections.” This second paper makes the novel hypothesis that the magnetosensitivity of the circadian system is related to a temperature-insensitive signalling system. The paper proposes an initial molecular model based on the Drosophila circadian clock, and provides a series of direct and inferential evidence in support of the hypothesis.


Introduction: Temperature insensitivity of circadian systems
Classical models of the circadian system that describe a negative feedback loop have the power to explain all the major features of the circadian system, but with one notable exception: the circadian system is insensitive to changes in temperature over a range of 10°C or more (1), a feature essential to circadian physiology. Whilst various proposals have been suggested for the temperature compensation of circadian clocks (1), the most widely discussed model involves interlocking-feedback loops (2). Such models predict that most mutations of circadian rhythm genes (which change kinetic constants and activation energies in the underlying control system) are likely to disrupt the delicate balances and to therefore exhibit failures in temperature compensation of the circadian rhythm. However, around 60–70% of circadian rhythm mutations result in intact temperature compensation, leading to some controversy over the notion that the robustness of circadian systems is the result of such “delicate balances” (1).

The existence of temperature compensation can be observed to have further effects of biological importance: 1) it plays a critical role in the coupling/synchronisation of peripheral clocks, which may receive only periodic zeitgeber signals from neuronal or endocrine inputs; and 2) it results in limited inter-individual variability in the circadian period, despite the fact that individuals show considerable genetic variability that translates into variations of kinetic parameters.

An alternate hypothesis – Circadian/GMF interactions are related to a temperature insensitivity mechanism  
The previous paper in this duo (3) established the repeated evidence for circadian/GMF interactions, but revealed that the existing explanatory framework – a secondary zeitgeber – is unsatisfactory. Instead, this paper reveals that there is existing evidence to suggest that temperature insensitivity and magnetosensitivity of circadian systems are related features.

Experimental studies on human subjects led Wever (4) to conclude that there was significant proof of the influence of the GMF on the human circadian system, but that the results were inconsistent with a secondary zeitgeber (see TABLE 1). Instead, it should be highlighted that under conditions of a null GMF, two key signatures of temperature sensitivity were observed: 1) decoupling between different rhythms within an individual and 2) an increase in inter-individual differences. Further experiments involved re-applying magnetic fields (MF), which successfully demonstrated reduced inter-individual variability and the re-establishment of internal synchrony in immediate response to the field. Whilst the results are inconsistent with a secondary zeitgeber effect (4), they are easily reconciled with the notion of the GMF being utilised by a temperature insensitivity device.

Furthermore, a study with Musca flies provides direct evidence of an apparent interaction between temperature, circadian activity, and the GMF (5). Again, the author concluded that the results were not evidence of a secondary GMF zeitgeber, but instead suggested that “one possibility is an influence on the coupling strength of oscillators”.

Table 1: Results of Wever (4), involving isolation experiments with human subjects, isolated from all cues of the 24h cycle. The results reveal a statistically different circadian period and greater inter-individual variability (SD) under shielded (e.g. null GMF) conditions. Most strikingly, internal desynchronisation (a decoupling of activity and temperature rhythms) – which is not expected in healthy adults – only ever occurred under shielded conditions. The final row reveals that apparent internal desynchronisation (e.g. one rhythm running with a period exactly double the other) only occurred under natural conditions. It can be speculated that the harmonic relationship between the two rhythms may be a signature of the underlying biophysical mechanism. Further experiments involved re-applying MFs, which successfully demonstrated the immediate re-establishment of internal synchrony and reduced inter-individual variability. It is further noted that CRY knockouts (e.g. see FIGURE 1 and (6,7)) mirror the above results with null MFs in many respects, with the observation of both increased SD and bi-component periodicities – e.g. an uncoupling of two different behavioural rhythms.

Parameter  Unshielded Room (n=34)  Shielded Room (n=50)  Statistical Significance 
Circadian Period 24.87h 25.26h p < 0.01
SD of Circadian Period +/- 0.44h +/- 0.85h p < 0.01
Internal desynchronisation 0 subjects 15 subjects p < 0.001
Apparent internal desynchronisation 5 subjects 0 subjects p < 0.01

Molecular model: A radical evolution of the circadian model 
In recent years, the “transcription translation feedback loop” (TTFL) model has undergone a fundamental revision with the recognition that there is core mammalian clockwork that is purely posttranslational, involving multimeric clock complexes comprising components including CRY, clock kinases and PER (a major phosphorylation target). Almost all clock components undergo maturation through sequential phosphorylation, which plays key roles in regulating degradation, transcriptional activity, complex formation, subcellular localization and activity (8). However, both models of the circadian system – a TTFL or a post-translation clock – would equally be expected to be at the mercy of reaction kinetics.

What is the role of cryptochrome within these multimeric phosphorylation clocks? Taking the Drosophila clock as our example, the role of cryptochrome is classically described to involve light-dependent resetting of the circadian clock. A light pulse early in the evening activates CRY, which triggers TIM for subsequent degradation by JET. As PER is reliant on TIM for stability, this results in a delay of PER accumulation and thus a delay in progression of the molecular clock. However, light also causes CRY degradation, but on a longer timescale– JET also binds light-dependently to CRY and promotes CRY’s degradation, but only after TIM is present at very low levels, thus allowing a new start of the circadian cycle (9). Conversely, a light pulse late at night degrades nuclear TIM, freeing PER to repress CLK/CYC activity earlier than normal, advancing timing of the onset of activity the next day.

The properties of the night-time phase response are thereby related to CRY/PER/TIM/JET interactions, in particular the subsequent degradation of CRY after TIM levels are almost absent. Thus, it is expected that the stoichiometric relationship of the various clock components are critical to the robustness and timing of the clock, as established experimentally in the analogous mammalian system (10). By logical extension, the robustness of the clock must furthermore be reliant on the relative pacing and/or stoichiometry of phosphorylation processes. Whilst the detailed mechanism of CRY signalling is presently unclear, the flavin dependent change is thought to lead to changes in the C-terminal tail (CTT) conformation, which gates formation of the proper target complex with its partners (e.g. JET/TIM) (11).

It is observed that whilst the CTT – responsible for interactions with partners (11) – is variable during evolution, the photolyase-like domain that is fundamentally responsible for forming radical pairs is highly conserved (3). What is the role of the photolyase domain in these multimeric clock complexes? One constraint that the radical-pair mechanism (RPM) has to overcome is that the energy of interaction of a molecule with the GMF is several orders of magnitude smaller than the average thermal energy. However, spin correlated radical pairs have the unique properties that their chemical fate is largely controlled by weak magnetic, rather than thermal, interactions (13). The RPM of CRY is therefore not only one of the few plausible biophysical candidates for magnetodetection: it is also one of the few plausible candidates for a temperature robust signalling mechanism.

It is therefore proposed that the role of CRY/RPM is to signal in a temperature insensitive manner to either pace phosphorylation processes, or alternatively to control the stoichiometry of phosphorylation processes (e.g. by switching equally between signalling states). Continuing with the Drosophila example, if CRY-induced TIM degradation was itself subject to reaction kinetics, then the properties of the clock would be temperature-sensitive. An initial model therefore proposes that CRY/RPM signalled degradation of TIM is the key temperature-insensitive reaction that ultimately sets the pace of the Drosophila circadian clock.

Figure 1: Results of Dolezelova et al. (6) and Kaushik et al. (14)

The vertical axis is the mean free-running circadian period in hours under constant conditions (constant dark unless otherwise stated); the horizontal axis is the experimental temperature in °C. These experiments directly implicate a role for cryptochrome in temperature insensitivity mechanisms. Dolezelova et al. (6) generated a full CRY knockout mutant (CRY0) revealing that these mutants behaved in an anomalously rhythmic manner in constant light. This is a distinctly non-wild-type behaviour, which occurs due to the lack of constant photic input (because CRY is absent) to the circadian system in such knockouts. Moreover, this circadian behaviour was revealed to be anomalously temperature sensitive. However, in constant darkness, the circadian system revealed a re-establishment of temperature insensitivity. This reveals two independent mechanisms of temperature insensitivity in
Drosophila – 1) a light/cryptochrome-dependent mechanism, and 2) a “dark”/cryptochrome-independent mechanism. The authors concluded that the results suggest a role for CRY beyond the classical Drosophila model, and that the gene is involved in core pacemaking. Further knockout studies (Kaushik et al. (14)) involved the long running period mutant perL alongside the loss-of-function cryptochrome mutant CRYb. The results revealed that whilst the CRYb retained temperature insensitivity, the CRYb mutant was capable of restoring the temperature-sensitive phenotype of perL mutants. The authors suggest that the interaction of CRY and PER-TIM are responsible for temperature insensitivity, and conclude that “interaction between CRY and PER-TIM complex is responsible for the loss of temperature compensation in the perL strain”(6).

Evidence for the model 
Evidence that temperature compensation mechanisms are embedded within multimeric protein complexes is derived from the clustering of temperature sensitive mutants, observed primarily within CRY and its partners e.g. PER and the clock kinases (see TABLE in (1)). However, there is generally an absence of such mutants in other components of circadian systems (1). In particular, CRY knockout Drosophila reveal anomalous temperature-sensitive circadian behaviour, but only under conditions of constant light (e.g. when CRY is active, see FIGURE 1) (6). The authors concluded that the results suggest a role for CRY beyond the classical Drosophila model, and that the gene is involved in core pacemaking. Further knockout studies led to a similar conclusion that temperature compensation is related to interactions between CRY and the PER-TIM complex (FIGURE 1) (14). Furthermore, both the Drosophila and mouse CRY knockouts (15,17) exhibit bi-component periodicities (e.g. the uncoupling of two different behavioural rhythms — a signature of temperature-sensitivity), with such findings thus mirroring the results of Wever with hypogeomagnetic fields on humans (4). Finally, it has recently been demonstrated that the phosphorylation status of Arabidopsis cryptochrome itself is sensitive to the intensity of the magnetic field (16).

It is proposed that CRY/RPM signalling is a fundamental component of the robustness and temperature insensitivity of circadian systems. The proposed model is not intended to replace the commonly held assumption that the temperature insensitivity of circadian systems is the result of a delicate balance of opposing influences (2). The two models are not mutually exclusive, but instead may contribute to the temperature insensitivity of circadian systems. Instead, it is highlighted that RPM/CRY is an ideal biophysical candidate for temperature insensitive signalling within the phosphorylation processes of multimeric post-translational circadian clocks. The biological outcome is to influence circadian behaviour in a manner that is independent of the influences of temperature and molecular environment on kinetic constants, thus further contributing to the robustness of circadian systems.

Whilst the hypothesis may be considered somewhat speculative, it does offer numerous advantages over the existing paradigm of a secondary zeitgeber in terms of both evidence and explanatory power, including: 1) the molecular biological mechanism is, at least for Drosophila, more clearly defined than the existing model (which remains entirely undefined); 2) it is more consistent with the established roles of cryptochrome and modern formulations of the post-translational circadian system; 3) it is mechanistically simpler and biophysically more plausible than the existing paradigm; 4) the proposed model provides a paradigm of clear biological utility across the animal kingdom; 5) it has the power to explain anomalous findings in the literature which are not all consistent with a secondary zeitgeber (e.g. internal desynchronisation of rhythms and increased SD under shielded conditions); 6) it has the power to explain the retention and purifying selection of a specifically configured magnetoreceptor with the circadian system through animal evolution. In fact, the model could even be applied beyond the animal circadian system, and explain the presence of a functional magnetoreceptor within timing processes in sessile species such as plants and corals; 7) whilst there is a plethora of negative evidence for the existing paradigm, the new hypothesis of temperature-GMF interactions is supported by evidence from Musca flies (5), indirect evidence from humans (4), and experiments revealing that temperature insensitive clock devices are indeed imbedded within cryptochrome-multimeric protein complexes (1,6,14).

However, extension of the hypothesis to mammalian species – where the core clock operates deep within the suprachiasmatic nucleus of the brain – makes the currently (12) unsupported assumption that mammalian CRYs are capable of light-independent routes of radical pair generation (however, this is not an unprecedented suggestion) (18). Moreover, the evidence for the hypothesis is presently supported by only a small number of experiments. Nonetheless, the core hypothesis is readily falsifiable – most notably, experiments investigating CRY-mediated GMF influences on the Drosophila circadian system (7) should be extended to more than a single temperature. H

Author declares no conflict of interest.


About the author 
James Close has previously held academic positions at King’s College London and the University of Oxford, where he applied his skills as a genome biologist to help unravel the biology of complex disorders in fields as diverse as haematology and psychiatry. He is now a freelance researcher and trader of financial markets.


  1. Hong CI, Conrad ED, Tyson JJ. A proposal for robust temperature compensation of circadian rhythms. Proc Nat Acad Sci. 2007 Jan 23;104(4):1195–200. 
  2. Ruoff P, Christensen MK, Sharma VK. PER/TIM-mediated amplification, gene dosage effects and temperature compensation in an interlocking-feedback loop model of the Drosophila circadian clock. J Theor Biol. 2005 Nov 7;237(1):41–57. 
  3. Close J. The Compass within the Clock – Part 1: The hypothesis of magnetic fields as secondary zeitgebers to the circadian system – logical and scientific objections. Hypothesis Journal. In Press. /10.5779/hypothesis.v12i1.359
  4. Wever R. ELF-effects on Human Circadian Rhythms. ELF and VLF Electromagnetic Field Effects [Internet]. New York: Publ. Plenum Press; 1974. p. 101–44. Available from: 
  5. Engelmann W, Hellrung W, Johnsson A. Circadian locomotor activity of Musca flies: Recording method and effects of 10 Hz square-wave electric fields. Bioelectromagnetics. 1996;17(2):100–10.<100::AID-BEM4>3.0.CO;2-7 
  6. Dolezelova E, Dolezel D, Hall JC. Rhythm defects caused by newly engineered null mutations in Drosophila’s cryptochrome gene. Genetics. 2007 Sep;177(1):329–45. 
  7. Yoshii T, Ahmad M, Helfrich-Förster C. Cryptochrome mediates light-dependent magnetosensitivity of Drosophila’s circadian clock. PLoS Biol. 2009;7(4):e1000086. 
  8. Tatarog˘lu Ö, Schafmeier T. Of switches and hourglasses: regulation of subcellular traffic in circadian clocks by phosphorylation. EMBO reports. 2010 Nov 5;11(12):927–35. 
  9. Peschel N, Chen KF, Szabo G, Stanewsky R. Light-Dependent Interactions between the Drosophila Circadian Clock Factors Cryptochrome, Jetlag, and Timeless. Curr Biol. 2009 Feb 10;19(3):241–7. 
  10. Lee Y, Chen R, Lee H, Lee C. Stoichiometric relationship among clock proteins determines robustness of circadian rhythms. J Biol Chem 2011 Mar 4;286(9):7033–42. 
  11. Zoltowski BD, Vaidya AT, Top D, Widom J, Young MW, Crane BR. Structure of full-length Drosophila cryptochrome. Nature. 2011 Nov 13; 480(7377):396–9. 
  12. Maeda K, Robinson AJ, Henbest KB, Hogben HJ, Biskup T, Ahmad M, et al. Magnetically sensitive light-induced reactions in cryptochrome are consistent with its proposed role as a magnetoreceptor. Proc Nat Acad Sci. 2012 Mar 14;109(13):4774–9. 
  13. Rodgers CT, Hore PJ. Chemical magnetoreception in birds: the radical pair mechanism. Proc Nat Acad Sci. 2009 Jan 13; 106(2):353–60. 
  14. Kaushik R, Nawathean P, Busza A, Murad A, Emery P, Rosbash M. PER-TIM Interactions with the Photoreceptor Cryptochrome Mediate Circadian Temperature Responses in Drosophila. PLoS Biol. 2007;5(6):e146. 
  15. Van der Horst GT, Muijtjens M, Kobayashi K, Takano R, Kanno S, Takao M, et al. Mammalian Cry1 and Cry2 are essential for maintenance of circadian rhythms. Nature. 1999 Apr 15;398(6728):627–30. 
  16. Xu C, Lv Y, Chen C, Zhang Y, Wei S. Blue light-dependent phosphorylations of cryptochromes are affected by magnetic fields in Arabidopsis. Adv Space Res [Internet]. Available from: 
  17. Vitaterna MH. Differential regulation of mammalian Period genes and circadian rhythmicity by cryptochromes 1 and 2. Proc Nat Acad Sci. 1999 Oct 12;96(21):12114–9.
  18. Ritz T, Adem S, Schulten K. A model for photoreceptor-based magnetoreception in birds. Biophys. J 2000 Feb;78(2):707–18.

Share your thoughts

Leave a Reply

You must be logged in to post a comment.