Biomedical Engineering Reference
In-Depth Information
The images of the dendritic tuft of a mitral cell (upper image -
low resolution; lower image - high resolution) are shown in (A).
Single pixel recordings of action potential signals from four indi-
vidual locations on the dendritic tuft in single-trial measurements
(no averaging) are shown in B. Action potential signals from
individual locations could easily be resolved. An improvement
in the signal-to-noise ratio with minimal spatial averaging (sig-
nals from the four pixels shown in B is illustrated in panel C). A
dramatic improvement with more extensive spatial averaging (the
whole tuft; 184 pixels) made the optical recording shown in D
appear similar in signal-to-noise ratio to electrode measurements.
Measurements of the same type were carried out during an evoked
synaptic potential at the dendritic tuft (
Fig. 3.4-I, E, F and G
).
Because excitatory post-synaptic potential (EPSP) was about 5
times smaller in amplitude than an action potential, we averaged
20 individual trials to obtain a signal-to-noise ratio similar to the
one corresponding to action potentials (the signal-to-noise ratio
increases with the square root of the number of averages). Panels
E, F and G of
Figure 3.4-I
again illustrate the striking improve-
ment in the sensitivity of recording with both spatial averaging
and averaging of individual repetitive events. This sensitivity was
routinely obtained from mitral cells that were positioned relatively
close to the surface of the slice (not deeper than 70
m) and that
had its primary dendrite and tuft close to the focal plane.
Figure 3.4-II
illustrates a representative example from
a series of optical measurements designed to determine the
characteristics of the evoked EPSP at the site of origin and its
attenuation along the primary dendrite. This information was not
possible to obtain before using electrode measurements because
of the small diameter of terminal dendritic branches. First, a
spike was elicited and optically recorded from multiple sites
in a single-trial measurement to serve as a calibration standard
(
Fig. 3.4-II B
, traces 1-10). The calibration of optical data from
multiple sites in terms of membrane potential requires a voltage
μ
Fig. 3.4. (continued) branch. (
C
) Calibration of optical signals (
F/F) in terms of membrane potential (mV). All traces
represent the average output of the same group of 35 pixels that receive light from the dendritic tuft (rectangles over the
tuft in A). Twenty trials were averaged to improve the signal-to-noise ratio. Trace 1 shows an optical signal corresponding
to an action potential of 93 mV used as a calibration standard. Trace 2 is a subthreshold EPSP signal evoked by olfactory
nerve stimulation and calibrated to be 9 mV in amplitude at the site of origin (tuft). Trace 3 shows a threshold EPSP signal
recorded from the tuft after the action potential was blocked by intracellular application of QX-314. In the measurement
shown in trace 3, the stimulus delivered to the olfactory nerve was identical to the one applied in the measurement shown
in trace 1. The EPSP signal in trace 3 (also superimposed over trace1) overlaps closely with the local response preceding
a spike in trace 1 indicating that the spike is eliminated by QX-314 while the synaptic potential was unchanged. (
D
) The
amplitude of EPSP signals on a voltage scale at 10 recording sites. The calibration of optical signals shows that EPSP at
location 8, only 15 μm away from soma, is 13.2 mV in amplitude. Electrical recording from the soma (trace 11 in (
D
))
was similar (12.5 mV). Traces
(9, 10)
are signals from left and right oblique dendritic branches.
