Literature1) Haag and Borst (2004) Neuronal mechanism underlying complex receptive field properties of motion-selective interneurons. Nat Neurosci 7:628-635
2) Beckers et. al. (2007) Synapses in the fly motion vision pathway: evidence for a broad range of signal amplitudes and dynamics. J Neurophysiol 97:2032-2041
3) Haag and Borst (2008): Electrical coupling of lobula plate tangential cells to a heterolateral motion-sensitive neuron in the fly. J Neurosci 28(53):14435–14442
To assess how the occurrence and timing of V1 spikes is affected by VS spikes we monitored V1
spiking activity when single VS spikes were elicited by brief current injections (H). V1 spikes were
tightly coupled to the injected current pulses in time (I). On occurence of a VS spike a V1 spike fol-
lowed with much higher probability than when the current pulse remained subthreshold, i.e. when it
failed to elicit a VS spike (J). The larger a VS spike is, the higher is the probability that a V1 spike is
triggered (K).
Voltage clamping of one VS
cell during rest (i.e. in the ab-
sence of visual motion) left
the postsynaptic spike rate
unaffected. However, the in-
terspike interval distribution of V1 spikes was affected. While in the non-clamped situation (L) many
doublet V1 spikes (marked by the X) occurred, the occurence of doublet V1 spikes was significantly
reduced when one VS cell was voltage-clamped (M, N).
When a presynaptic VS cell is voltage clamped V1 spikes are accompanied by a brief hyperpolari-
zing current transients in the clamped VS cell (C). Some V1 spikes (3.1%) are not accompanied by
a detectable current transient (D-E). The amplitu-
des of these transients are not uniform (F). Some
VS2/3 cells showed a clear bimodal distribution of
current transient amplitudes (G).
The occurence of ‘failures’ (V1 spikes not accompanied
by a current transient in VS), the different amplitudes
of the current transients, and in particular the bimo-
dal distribution, speak against the assumption that the
observed current transients result mainly from electri-
cal coupling between VS and V1. We therefore propo-
se that due to electrical coupling among the VS cells
the current transients are caused by the occurrence
of spikes in neighboring VS cells.
We conclude the spike rate of the postsynaptic V1-neuron to be controlled by both, graded voltage
modulations and spikes of the presynapic VS cells. Whereas the graded presynaptic component re-
gulates the postsynaptic spike rate to a large extent, the presynaptic spikes predominantly control
the actual timing of postsynaptic spikes. Presynaptic spikes play a role for ensemble synchroniza-
tion, as is evident from the decrease in the rate of postsynaptic spike doublets when one VS cell is
prevented from spiking. Signal transmission from VS to V1 is largely mediated by chemical synap-
ses, but an additional contribution from electrical coupling cannot be excluded and an eleclectrical
coupling between VS1 and V1 has recently been shown3.
Introduction Results
Results
Conclusions
Ulrich Beckers, Martin Egelhaaf and Rafael KurtzDepartment of Neurobiology, Bielefeld University, 33501 Bielefeld, [email protected]
20 ms50 ms
1 nA
50 ms
D
E
1 nA
0 20 40 60 80 100 >1200
10
20
30
40
50
60
70
80
90
100
110
0
0.05
0.1
0.15
0.2
0.25
0 20 40 60 80 100 >1200
10
20
30
40
50
60
70
80
90
100
110
vcmode
bridged
num
ber
num
ber
V1 s
pike
dou
blet
pro
babi
lity
interspike interval [ms] interspike interval [ms] mode
200 ms
10 mV
200 ms
10 mV
ML N
Information transfer between synaptically coupled neurons
can be mediated by spikes, by graded signals, or by combi-
nation of both types of signals. We simultaneously recorded
from identified visual motion sensitive cell pairs of the fly (A).
The presynaptic VS cells convey both graded and spike si-
gnals (VS membrane potential: blue traces) to their common
purely spiking postsynaptic target cell V1 (V1 spikes indicated
as red dashes). The VS cells are electrically coupled among
one another1. Synaptic transmission of graded signals from
VS to V1 is linear over a broad signal range2. Postsynaptic
spikes were elicited by voltage clamp (C) or injecting brief
current pulses (H) into one VS cell. In this study we analyzed
the time course of the holding current (C-G) to monitor the input received by the presynaptic cell and
the correlation of V1 spike activity to single VS spikes (H-K). The role of fast signals for postsynaptic
synchronization was studied by functionally removing one VS cell from the input ensemble (L-N).
Contribution to the 8th NWG Göttingen Meeting 2009
Impact of presynaptic voltage transients on postsynaptic spike
timing at a graded synapse in the fly motion vision system
intracellular recording electrode in VS axon
extracellularrecording electrodein V1's output region
registeredspikes
VS-V1 synapse
null direction preferred direction
resting potential
VS cell retinotopic input region
V1
VS1VS2VS3VS4
A
B
C
10 mV
50ms
10 mV
100 ms
100 ms
1 nA
20 mV200 ms
1 nA
0 -2 -30
60
120
180
0 -1 -2 -30
40
80
120
0 -1 -2 -30
5
10
15
0 -1 -2 -30
10
20
30
0 -1 -2 -30
15
30
45
F
Gall 0 mV 10 mV
30 mV 50 mV
-1current [nA]
num
ber
num
ber
current [nA]
#1 #2 #3 #4 #5 #6 #7 #8 Sum0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1I
prob
abili
ty
cell
2nA
50ms
10mV
H
114
246
25
195
209
631
185
695
140
380
78
342
109
391
59
3141
919
261
0 10 30200
1
2
3
40 50time [ms]
V1
spik
e ra
te [r
elat
ive
units
]
J K
0 0.5 1 1.5 2 2.5 3 3.5 4 x0102030405060708090
100
VS
-V1
spik
e co
uplin
g ra
te [%
]
relative threshold [mV]
n=21
n=10n=11
n=11
n=7
n=6
n=3
n=7
n=21
n=6