Abstract
Honey bees (Apis mellifera) offer an alternative model for investigating magnetoreception, exhibiting reliable navigation and behavioral responses to magnetic fields. Thanks to their compact brains, well-matched to the penetration depth of modern optical imaging techniques, they may offer insights into the neural and biochemical mechanisms underlying this sense, something that standard models, like migrating birds, have not so far provided. But also in honeybees, this progress requires tools capable of resolving weak, magnetically induced neural activity with high spatio-temporal precision. The approach, presented here, bridges quantum biology and neuroscience, aiming at testing the radical pair mechanism (RPM) as a potential basis for magnetic sensing. As the RPM predicts that magnetoreception is coupled to the visual system, we developed an in vivo two-photon calcium imaging approach to measure neural activity in the anterior optic tubercle, a higher-order visual center involved in chromatic processing and potentially navigation. Bees were prepared using a minimally invasive technique, in which this neuropil was retrogradely labelled with a fluorescent calcium indicator, enabling stable recording conditions over several hours. Controlled blue-light stimuli were provided by the scattered output of a fiber laser, and weak magnetic-field stimuli were applied by a shielded, three-axis Helmholtz coil system that allowed precise modulation of field strength and polarity while minimizing electromagnetic interference. Visual stimulation evoked consistent and reproducible calcium responses, validating the preparation and imaging stability. Magnetic stimulation produced small decreases in the calcium-response signal in single individuals without statistical significance, which might be the first indications of field-dependent modulation of neural activity. The developed imaging framework will allow thorough testing of magnetic modulation in vision- and navigation-related brain regions, providing a platform for controlled tests of RPM-specific predictions, including light dependence, polarity independence, and radiofrequency perturbation. By targeting other brain regions, the platform can also be adapted to test alternative mechanisms beyond the visually mediated RPM hypothesis, like putative magnetite-based sensing in the bee's abdomen, as implicated by previous studies.