Speaker
Description
The light-sensitive flavoprotein CRYPTOCHROME (CRY) is now generally believed to be a magnetosensor through the Radical Pair Mechanism (RPM). In this model a light-initiated electron transfer between CRY’s co-factor Flavin Adenine Dinucleotide (FAD) and a chain of essential tryptophan (Trp) residues provides sensitivity to a magnetic field (MF) via a quantum effect on electron spin, ultimately modulating CRY protein activation and downstream signalling. Whilst there is considerable physical and behavioural data to support this Trp-FAD canonical RPM, particularly as the mechanism for navigational magnetoreception in birds, a growing body of evidence suggests that non-canonical radical pairs may underlie other forms of magnetosensitivity.
Using the model organism Drosophila melanogaster, renowned for its highly tractable nervous system and genetic malleability we investigated the molecular basis of CRY based magnetosensitivity. Using a combination of electrophysiological and behavioural assays we initially demonstrated that just the 52 amino acids of the C-terminus of Drosophila CRY (Dm-CT) alone were capable of imparting magnetosensitivity to a fly. Lacking in both the chain of Trp residues as well as the entire FAD binding domain this surprising result cannot be explained by the canonical RPM. By manipulating the intracellular concentration of flavins such as FAD and its metabolic precursor Riboflavin we demonstrate that intramolecular radicals from FAD and the Dm-CT interact to provide magnetosensitivity.
To investigate how Dm-CT and FAD interact we applied all-atom molecular dynamic (MD) simulations. These revealed that several key positively charged residues distributed across the Dm-CT mediate the direct binding of FAD via electrostatic forces in several possible configurations. As well as localising intramolecular FAD radicals to Dm-CT the MD simulations suggest that upon binding, the FAD molecule is contorted into a ‘open’ conformation, enhancing the sensitivity of the radicals to a MF.
Directed by these simulations we generated genetically modified flies with substitutions of the key positively charged residues for structurally conservative residues with negative and neutral charge. In both our electrophysiological and behavioural assays flies lacking the correct positive charge distribution within the CT failed to support magnetosensitivity. Complementary in vitro binding investigations supported our MD simulations and in vivo observations, demonstrating a physical interaction between Dm-CT and FAD through both size exclusion chromatography and fluorescence microspectroscopy. Notably, these investigations reveal an FAD and light dependent propensity for Dm-CT to form aggregates, reminiscent of photobodies found in plant CRYs.
Whether this non-canonical RPM has relevance to navigational magnetoreception or is a unique quirk of Drosophila remains to be determined. With accumulating evidence for non-canonical magnetosensitivity, such as that of Human CRY2 (HsCRY2) which despite a defunct FAD binding domain still exhibits light and MF sensitivity in Drosophila, we present preliminary evidence suggestive of a possible similar C-terminus – FAD interaction mediating magnetosensitivity for HsCRY2 in Drosophila.
Finally, we explored the dual role of Dm-CT as not just a facilitator of FAD radical formation, but as a transducer of the magnetic signal. Using our electrophysiological and behavioural assays we demonstrate the requirement for protein-protein binding sites for the magnetic signal and evidence that implicates possible signalling candidates in the transduction of the magnetic sense in flies.