The role of the AFD neuron in C. elegans thermotaxis analyzed using femtosecond laser ablation

Abstract

Background: Caenorhabditis elegans actively crawls down thermal gradients until it reaches the temperature of its prior cultivation, exhibiting what is called cryophilic movement. Implicit in the worm's performance of cryophilic movement is the ability to detect thermal gradients, and implicit in regulating the performance of cryophilic movement is the ability to compare the current temperature of its surroundings with a stored memory of its cultivation temperature. Several lines of evidence link the AFD sensory neuron to thermotactic behavior, but its precise role is unclear. A current model contends that AFD is part of a thermophilic mechanism for biasing the worm's movement up gradients that counterbalances the cryophilic mechanism for biasing its movement down gradients.

Results: We used tightly-focused femtosecond laser pulses to dissect the AFD neuronal cell bodies and the AFD sensory dendrites in C. elegans to investigate their contribution to cryophilic movement. We establish that femtosecond laser ablation can exhibit submicrometer precision, severing individual sensory dendrites without causing collateral damage. We show that severing the dendrites of sensory neurons in young adult worms permanently abolishes their sensory contribution without functional regeneration. We show that the AFD neuron regulates a mechanism for generating cryophilic bias, but we find no evidence that AFD laser surgery reduces a putative ability to generate thermophilic bias. In addition, although disruption of the AIY interneuron causes worms to exhibit cryophilic bias at all temperatures, we find no evidence that laser killing the AIZ interneuron causes thermophilic bias at any temperature.

Conclusion: We conclude that laser surgical analysis of the neural circuit for thermotaxis does not support a model in which AFD opposes cryophilic bias by generating thermophilic bias. Our data supports a model in which the AFD neuron gates a mechanism for generating cryophilic bias.

Figure 1

Femtosecond laser ablation. A. Diagram of laser surgery setup. Amplified femtosecond laser pulses are tightly focused by an objective lens (1.4 NA) to dissect targets positioned with a three-dimensional piezoelectric stage and visualized by fluorescence microscopy. B. DiO-stained amphid dendrites before and after severing the middle dendrite using 3-nJ pulses without visibly affecting neighboring dendrites as close as 500 nm away. C. An example of a GFP-labeled AFD dendrite retracting after surgery using 6-nJ pulses. D. Confocal microscope image showing GFP-labeled AFD neurons the day after surgically severing one dendrite. The position of the cut along the dendrite is representative of all experiments. E. Phasmid DiO (pre-surgery) and DiI (post-surgery) staining 24 hours after surgery of the PHA dendrite. The PHA cell body does not absorb DiI because it is physically disconnected from its sensory endings. The cut in the dendrite is indicated with the arrow.

Figure 2

Microdroplet assay. A. Sample data of cryophilic bias at temperatures above T_cult. The top panel shows the sinusoidal thermal cycle recorded at the microdroplet. Sample video frames depict worm conformation and the corresponding body extension; dips of body extension below a threshold are counted as reorientation events. A time series of reorientation events is presented in the fourth panel as a raster of all cycles. The bottom panel is a histogram of reorientation events, showing the distribution of reorientations with respect to the thermal cycle. To avoid sensitivity to possible behavioral phase delays, we use the data from the 10 s in the middle of warming and cooling intervals (dark histogram bars). The tendency to reorient more frequently during the warming phase than the cooling phase of the thermal cycle reveals cryophilic bias. B. Sample data of unbiased movement below T_cult. Reorientations occur randomly throughout the thermal cycle, showing that this worm is not exhibiting thermophilic or cryophilic bias.

Figure 3

Effects of laser surgery on cryophilic bias. The amount of cryophilic bias exhibited by worms assayed above and below T_cult. In each panel, black bars correspond to experiments with worms that did not undergo surgery or surgical preparation, gray bars correspond to experiments in which AFD dendrites were severed, and white bars correspond to experiments in which neurons were killed. Mock control measurements represent worms that were prepared for surgery without undergoing laser irradiation. The values of the thermotactic index are means ± one standard error. The cases in which surgery significantly affects cryophilic bias in comparison to its respective mock control are indicated by asterisks (***, P < 0.0005; *, P < 0.05). All other cases are not distinguishable from their respective mock controls (P > 0.2). Significance was measured using a two-tailed Student's t-test. The number of worms tested in each measurement of cryophilic bias is noted next to each data bar.

Figure 4

A gating model for AFD and AIY in generating cryophilic bias. In this schematic, the synaptic connections between AFD and AIY and between AIY and AIZ are inhibitory, and AIZ contributes directly to generating cryophilic bias. We suggest that patterns in the activity of the neural circuit differ when T > T_cult and when T <T_cult. High and low levels of activity are differentiated by black and gray, respectively.