The role of horizontal system cells in optomotor responses in Drosophila melanogaster

When confronted with a large-field stimulus rotating around the
vertical body axis, flies display a following behaviour of the head
and steer in the direction of motion. As neural control elements
for this so-called 'optomotor response', the large tangential
horizontal cells (HS-cells) of the lobula plate have been the prime
candidates for long. When HS-cells are surgically damaged or
genetically removed, flies display reduced optomotor responses. To
provide a better understanding of the role of HS-cells in the
control of optomotor behaviour three approaches were taken. First,
experiments were designed to investigate which of the HS-cells
could be participating in head yaw movements in fixed flies and yaw
turning behaviour during tethered flight. Horizontal motion at
different elevations was presented to the flies. Comparison of the
optomotor responses with HS-cell receptive fields suggests that HSN
and HSE participate in head yaw movements whereas all three
HS-cells are used to control yaw turning behaviour during flight.
Second, to test whether HS-cells are sufficient to elicit yaw
optomotor responses, a bi-stable Channelrhodopsin-2 variant 'ChR2
(C128S)' was expressed in HS-cells using the Gal4 / UAS-system.
Combining a blue light stimulus with ChR2(C128S) allowed to
activate HS-cells without presenting a visual stimulus to the eye
of the fly. These experiments revealed that blue light was
sufficient to evoke robust head yaw movement in fixed flies as well
as turning behaviour in tethered flying flies, thus, mimicking
front-to-back visual stimulation on the stimulated side. Third, the
role of the receptive field layouts of HS-cells for optomotor
responses was studied. Flies with a gain-of-function of a single
Dscam1 isoform in all HS-cells (Dscam gain-of-function (D(GOF))
flies) were tested. Compared with HS-cells of control flies,
HS-cells of D(GOF) flies show reduced sensitivity to horizontal
motion in the frontal and enhanced sensitivity to motion in the
lateral part of visual space. The optomotor response of tethered
flying flies were analyzed. Compared with control flies, D(GOF)
flies responded significantly weaker to visual stimuli extending
over the entire azimuth extension of HS-cell receptive fields.
Stimulating flies with additional motion in the rear part of visual
space significantly reduced optomotor responses of control flies,
whereas (GOF) flies responded to both visual stimuli with about
equal strength. Although D(GOF) HS-cells had dramatically reduced
sensitivity to motion in the frontal part of visual space, D(GOF)
flies responded robustly to motion in this region of visual space.
D(GOF) and control flies also showed differences in head yaw
movements. These behavioural differences did not correlate with the
difference in the receptive fields of D(GOF) and control HS-cells.
The experiments indicate that HS-cells are sufficient to trigger
yaw turns of the head and whole body. All three HS-cells control
body turns during flight and head yaw turns are controlled by HSN-
and HSE-cells. The layout of HS-cell receptive fields, however,
does not correlate 1 : 1 with optomotor responses. During flight,
flies rely additionally on cells sensitive in the frontal part of
visual space. Furthermore, the layout of the HS-cell receptive
fields is important for incorporating motion information in the
rear part of visual space to the optomotor responses.