The role of identifiable neurons in magnetic orientation of the sea slug
Tritonia diomedea.

S.D. Cain, J. H. Wang, K. J. Lohmann
University of North Carolina at Chapel Hill, Dept of Biology


INTRODUCTION

Many animals can detect the Earth's magnetic field and use it as a cue in orientation and navigation. However, the neural mechanisms that underlie magnetoreception have remained elusive. The nudibranch mollusc, Tritonia diomedea, possesses two individually identifiable neurons that respond to changes in earth-strength magnetic fields1. These neurons, left pedal 5 (LPd5) and right pedal 5 (RPd5), are the only individually identifiable neurons known to respond to changes in earth-strength fields. As the two cells are located in similar areas of the symmetrical, paired pedal ganglia and have the same morphological and electrical characteristics, they are considered bilaterally symmetrical homologs. This study investigates the role of these two neurons in the neural circuitry involved in the magnetic orientation behavior of Tritonia.

The Pd5 cells have large (~ 500 µm), whitish cell bodies located in the two pedal ganglia (Fig.1). Electrophysiological studies of Willows and colleagues2 indicated that the pedal ganglia both receive input from and provide output to the ventral foot area. It is therefore possible that these cells can have one or more functions in the magnetoreception circuitry. For example, the Pd5 neurons could function as a primary receptor cell, interneuron, motor neuron, or modulatory neuron.

In Tritonia, the beating of pedal cilia is the primary mode of locomotion. The cilia propel the animal over a layer of viscous mucus secreted by the foot epithelial cells. Willows and Lloyd3 isolated three related novel neuropeptides (TPeps) that increase ciliary transport rate when applied to isolated patches of ciliated foot epithelium. Similarly, TPeps applied to isolated ciliated foot epithelial cells resulted in an increase in ciliary beat frequency. The peptides are present in the cell body and neurites of Pd5 cells. These results, combined with the location and morphology of the Pd5 neurons, suggest that one function of these cells is to modulate locomotion and turning during magnetic orientation behavior.

To determine if this is indeed the case, we investigated the anatomy of the pedal nerves containing axons of the Pd5 neurons. These axons would necessarily innervate the foot epithelia. We also determined the direction of action potential propagation indicating whether information is being sent toward or away from the CNS. Last, we used immunocytochemistry to determine if axons containing TPeps are located adjacent to the ciliated epithelial cells.

Fig. 1

The sea slug, Tritonia diomedea

Fig. 2

The fused ganglia of the Tritonia brain. This study focused on the large, whitish Pd5 neurons and the two nerves, PdN2 and PdN3,that contain axons projecting from these cells.

INVESTIGATIONS AND RESULTS

A. Anatomy of pedal nerves 2 & 3-

Two nerves innervating the pedal ganglia, PdN2 and PdN3, contain axons of the Pd5 neurons. If Pd5 is involved in paracrine signaling to cilia, the two nerves must directly innervate the foot epithelium. Therefore, methylene blue stain was used to ascertain the gross anatomy of the two nerves.

Fig. 3

Cobalt-fill of LPd5. The neuron projects axons through left pedalnerve 2 (LPdN2) and left pedal nerve 3 (LPdN3).

Fig. 4

Schematic showing the gross innervation pattern of LPdN3.The nerve leaves the brain and continues ventrally. The nerve branches upon reaching the lateral body wall, the main branch continuing posteriorly. The nerve branches repeated as it continues posteriorly, sending collaterals into the ventral body wall directly adjacent to the ciliated foot epithelium.

B. Direction of action potential propagation in Pd5-

Pedal nerves PdN2 & PdN3 are mixed nerves containing axons of both sensory and motor neurons. Therefore, Pd5 could be either receiving input or providing output through these nerves. Using electrophysiological techniques, the direction of action potential propagation in Pd5 neurons was determined. Intracellular recordings from the cell body of LPd5 and extracellular recordings from either LPdN2 or LPdN3 were obtained simultaneously. By comparing the time of action potential initiation in the cell body with that of the corresponding potential in the nerve, the direction of propagation was discerned.

Fig. 5

Electrophysiological recordings showing the direction of action potential propagation of neuron LPd5 though LPdN2 A. Evoked potentials in both the cell body and LPdN2. The action potentials in the cell body show a 1:1 correlation with the large potential recorded from LPdN2. Inset shows the time course of the two potentials recorded simultaneously. B. Spontaneously occurring activity in LPd5 and LPdN2. The action potential reaches the cell body prior to reaching the extracellular recording site. Inset shows continued 1:1 correlation of potentials in the two recordings.

Fig. 6

Electrophysiological recordings showing the direction of action potential propagation of neuron LPd5 though LPdN3 A. Evoked potentials in both the cell body and LPdN3. The action potentials in the cell body show a 1:1 correlation with the large potential recorded from LPdN3. Inset shows the time course of the two potentials recorded simultaneously. B. Spontaneously occurring activity in LPd5 and LPdN3. The action potential reaches the cell body prior to reaching the extracellular recording site. Inset shows continued 1:1 correlation of potentials in the two recordings.

C. Immunolocalization of TPep in the ciliated foot epithelium-

Immunohistochemistry has shown that TPep-like immunoreactivity is present near the ciliated foot epithelium4. To determine if the TPep is acting in paracrine signaling to the ciliated cells, immunocytochemistry was used to localize TPep-like immunoreactivity in the foot epithelium. Epithelial tissue was fixed in 4% paraformaldehyde; 0.5% glutaraldehyde using microwave technology. Tissues were labeled with a primary rabbit antibody against TPep-NLS, followed by an antirabbit, horseradish peroxidase congugated or gold congugated secondary antibody.

Fig. 7

Light level immunogold-silver enhancement of TPep immunoreactivity in Tritonia. A. Labeling in LPd5 with the primary antibody (+) and without the primary antibody (-). B. Labeling of medial and lateral ciliated foot epithelium probed with the primary antibody. Both show the localization of TPep immunoreactivity similar to that observed by Willows et al.4

Fig. 8

Immunocytochemistry of TPep immunoreactivity. DAB staining is evident in nervous structures adjacent to the basement membrane of the epithelium. The staining is also present in nervous structures that have penetrated the basement membrane and are directly juxtaposed to the ciliated cells.

SUMMARY

Our study furthered the understanding of one function of the left and right pedal 5 neurons during magnetic orientation behavior. The findings suggest the Pd5 neurons are able to modulate the beat frequency of ciliated cells of the foot epithelium. The cells possess axons that project out of two nerves that innervate the area near the ciliated cells. In both nerves, the Pd5 axons appear to be sending information out of the CNS and to the foot epithelium. Additionally, the TPep neuropeptides, which are known to affect ciliary beat frequency in isolated cell preparations, are present in nerves adjacent to the basement membrane of the foot epithelium. The localization of the neuropeptides to nerves near the ciliated cells implies that these nerves have the potential to directly alter the rate of ciliary beating. Therefore it is likely that these cells (Pd5 and Pd6) are involved in Tritonia's motor response during orientation behavior.

LITERATURE CITED

1 Lohmann, K.J., Willows, A.O.D., and Pinter, R. (1991). J exp Biol. 161, 1-24.
2 Willows, A.O.D., Dorsett, D.A., and Hoyle, G. (1973). J Neurobiol. 4, 207-238.
3 Lloyd, P.E., Phares, G.A., Phillips, N.E., and Willows, A.O.D. (1996). Peptides 17, 17-23.
4 Willows, A.O.D., Pavlova, G.A., and Phillips, N.E. (1997). J exp Biol. 200, 1433-1439.

ACKNOWLEDGMENTS

This work was supported by NSF Research Grant IBN- 9631951 to K. J. L. We thank Drs. A. Willard and W. Kier for helpful discussions and suggestions regarding the research project. We also thank Dr. R. Bagnell and Ms. V. Madden for the use of the UNC Pathology Microscopy Facility, and A. O. D. Willows for supplying the TPep antibodies.