Transcript Document

William Buchser
Neuroscience
Journal Club
April 28th 2004
Timing of neural responses in
cortical organotypic slices
Background
Timing
• Events which occur with timing greater than a
few seconds is probably grasped by out frontal
cortex – a component of complex thought
processing.
• Timing in the shorter range (less than a second or
two) is thought of as automatic – the brain has a
system that deals with it unconsciously.
“The neural representation of time.” Ivry RB, Spencer RM
Background
Ordinal
Metric
WORD
L R
Duration
Discrimination
Parkinson’s Disease
“The neural representation of time.” Ivry RB, Spencer RM
Timing Examination
400 ms
L
R
L
R
200 ms
600 ms
300 ms
http://www.neurobio.ucla.edu/~dbuono/InterThr.htm
Background
• Our temporal resolution in the auditory system is
much higher than in other sensory systems
• Timing has been studied using psychological,
behavioral, and neuro-imaging approaches.
• Few in vitro studies exist  this paper!
• Can we observe a physiological ‘memory trace’ in
brain slices?
“The neural representation of time.” Ivry RB, Spencer RM
This paper uses organotypic
slices to try determine if
reliably timed ‘late’
responses can be observed in
vitro, and what mechanisms
lead to their existence.
Methods
- Adult Sprague Dawley Rats
- Remove Brain and make slices
- Use coronal slice, 350400μm thick slices containing
auditory cortex
http://caddlab.rad.unc.edu
Organotypic Slices
Coronal slices are
placed on special
membranes used for
culture, they are bathed
constantly in artificial
cerebrospinal fluid.
Millipore: Millicell®-CM
Recording Paradigm
A single pulse was applied every 15-30 s to elicit synaptic responses.
Stimulation intensity ranged from 30 to 100 µAmpere.
Stimulating
Electrode
Recording
Pipette
Dorsal
Ventral
Lateral
Medial
Slice from: Redefining the tonotopic core of rat auditory cortex: Physiological evidence for a posterior field
Neot N. Doron, Joseph E. Ledoux, Malcolm N. Semple
Horizontal Voltage Sweeps
Timing of the Late Response
-50mV
Stimulus
Average response of
60 trials below
-80mV
Figure 1A
Timing of the Late Response
-30mV
-80mV
Time (min)
Action Potentials
Figure 1B
Timing of the Late Response
• We now know what the ‘late’ response looks like.
• A monosynaptic PSP is observed within the first
few milliseconds after stimulus.
• After more time (between 50 and 400 ms) more
EPSPs occur (probably polysynaptic in nature), and
are sometimes integrated into action potentials (all
of this is the ‘late’ response).
Summary - Figure 1
What does mono and polysynaptic mean?
Stimulus
Latency vs. Accuracy
•What is the latency and accuracy of the timed
responses?
• By examining the mean and the SD for the
latency of the first spike, we can make some
conclusions about relationships therein.
Preface - Figure 2
Mean latency of the first spike versus the SD
Correlation
r = 0.94
High Variability
Long Latency
High Accuracy
Short Latency
Figure 2
Latency vs. Standard Deviation
• Strong correlation between the latency and the
SD (accuracy) of firing.
Summary - Figure 2
Electrode Distance and First Spike Latency
• Does the distance from the stimulating electrode
correlate to the observed latency?
• Others have made the observation that with a
monosynaptic connection, distance is highly
correlated with latency.
•Polysynaptic connections on the other hand, may
or may not be correlated with spatial relationships.
Preface - Figure 3
Recording Paradigm
650µm 1200µm
Stimulating
Electrodes
Recording
Pipette
Dorsal
Ventral
Lateral
Medial
Figure 3
A cell that fires sooner to more distant stimulation
Figure 3B
Distance vs. Mono and Poly Response Latency
Figure 3A
Electrode Distance and First Spike Latency
• Distance from stimulating electrode affects the
latency of the sub-threshold monosynaptic
response.
• Distance from the electrode has no simple
relationship with the latency in neuron firing
(polysynaptic integration leading to response)
Summary - Figure 3
The Network nature of the Late Response
• The complex nature of the ‘late’ response of a
single cell implies polysynaptic network input.
• To test this, we will determine the extent that the
late response is reliant on NMDA receptors.
• We will use the NMDA blocker APV to determine
if NMDA receptors play a role in the late response.
Preface - Figure 4
Late responses are dependent
on NMDA receptors
-10mV
-80mV
Figure 4A
Network nature of the Late Response
• NMDA receptors are indicated in the late
response due to an abolition of polysynaptic
potential during APV’s application
• Therefore, neurons are likely to get excitatory
polysynaptic input which needs to be integrated
before a supra-threshold response occurs.
Summary - Figure 4
Paired Recordings
• To determine whether different cells exhibit
similar responses to the same stimulus, we
recorded simultaneously from two neurons.
Preface - Figure 5
Recording Paradigm
50µm
750µm
Stimulating
Electrode
Recording
Pipettes
Dorsal
Ventral
Lateral
Medial
Figure 5
Different cells respond to same stimulus
750µm
50
Time (min)
350
Figure 5A
Average sub-threshold response of two cells
Red cell firing elicited an EPSP in blue cell
AP
Stimulus
AP
750µm
Figure 5A
Two cells compared by correlating their
sub-threshold output.
Mean = 0.84
Figure 5C
Paired recordings and shared input
• Figure 5 demonstrates that two different neurons
have different responses to the same stimulus
• Although the action potentials are different, a lot
of the subthreshold response is similar between
them, reflecting shared input from the network.
• Regardless of shared inputs, there is significant
difference such to produce different suprathreshold responses.
Summary - Figure 5
Cross-Correlation between two neurons
Figure 6A
First-spike latency between neurons that
fire at different intervals is correlated.
Figure 6B
Other correlations of paired recordings
• The scatter plot indicates that there is a strong
correlation between the first spike in the red and
green neuron
• Activity in the green cell is contributing to the
firing of the red cell.
Summary - Figure 6
Stimulus intensity and first spike latency
• To find if hard-wired ‘delay lines’ exist in the
network, we can vary different stimulus parameters
and see if the latency changes.
• For the final experiment, they test whether
intensity has an effect on the latency of the late
response
Not in Printed Copy
Preface - Figure 7
Stimulus intensity and first spike latency
-20mV
High
Low (5-10% below high)
-80mV
Figure 7A
Stimulus intensity and first spike latency
Figure 7B
Stimulus intensity and first spike latency
• Latencies are stimulus-level (intensity) dependent.
• Therefore the latency is not hardwired into a
particular neuron.
• Stimulus intensity may be involved in the
mechanism in which timing is learned.
Not in Printed Copy
Summary - Figure 7
Summary
• Action potentials occurring up to 300 ms after
stimulus onset are observed in organotypic slices
from rat auditory cortex.
•These late responses are attributed to dynamics of
the neural network
•Buonomano hypothesizes that timing in this range
could be a local process, reflected in the dynamics
of neurons recruited in a task-specific manner.
“The neural representation of time.” Ivry RB, Spencer RM