Sorting the connections with multi- electrode neuronal ensemble recording techniques: Single-unit and local field potential activity from rat to man Dr Rob Mason Institute of Neuroscience School of Biomedical Sciences University of Nottingham Medical School and the LAB TEAM uronal Networks Laboratory
54
Embed
Sorting the connections with multi-electrode neuronal ensemble recording techniques:
Neuronal Networks Laboratory. Sorting the connections with multi-electrode neuronal ensemble recording techniques: Single-unit and local field potential activity from rat to man Dr Rob Mason Institute of Neuroscience School of Biomedical Sciences - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Sorting the connections with multi-electrode neuronal ensemble recording
techniques:
Single-unit and local field potential activity from rat to man
Dr Rob Mason
Institute of Neuroscience
School of Biomedical Sciences
University of Nottingham Medical School
and the LAB TEAM
Neuronal Networks Laboratory
LAB TEAM
epilepsy Ben Coomber [Dr Mike O’Donoghue ~ Dept
Neurology, QMC]
Clare Roe schizophrenia Dr Jill Suckling [Prof CA Marsden]
Dr Dissanayake anxiety Dr Carl Stevenson [Prof CA Marsden]
pain Dr Steve Elmes [Dr V Chapman]
rodent USVs Beth Tunstall [Dr S Beckett]
data analysis Margarita Zachariou [Prof S Coombes / Dr M Owen]
[Mathematics Dept]
Dr David Halliday (University of York)Prof D Auer (fMRI / phMRI ~ QMC)
• Rat language ~ ultrasound vocalisations (USVs) & affective state
Neuronal Networks Laboratory
Sorting the connections with multi-electrode neuronal ensemble recording techniques
To & From ~ rat prefrontal cortex
Sorting signal from noise
Ensemble Neuronal Unit activity
LFP activity
DATA Analysis & Interpretation
Distribution of neurones contributing to signals recorded by tetrode array
after Busaki 2004
Bionics 100-channel (“hedgehog”) array
Michigan MEAprobe 16-channels
NBLabs 8-channel array
4-channel - independent manipulation
Examples of Electrode Arrays
MULTIPLE ELECTRODE ARRAY RECORDING in vivo
• Single-site recording
e.g. hippocampal sub regions (CA1 & CA3)
• Multiple-site recording
e.g. prefrontal cortex & hippocampus
NBLabs 16-channel micro-wire array
• CA1 & CA3
Multiple Electrode Recordings in vivosimultaneous multichannel spike & field potential
recording
• 64-channel MAP system• 32-channel MAP system – with CinePlex for behavioural studies • 64-chanel Recorder system• 16-chanel Recorder system used MAP & Recorder for in vitro MEA studies with brain slices & neuronal cultures
Extracellular Recording - signal filtering & discrimination
AP Spike discrimination: separate action potential (AP) signal from noise
• AP amplitude detection / AP waveform shape recognition
Signal filtering: separate unit activity and Local Field Potentials (LFPs)
0 0.10 0.20 TIME (s)
Raw
Sig
nal
(V
) 0
100
100
0Filt
ered
Sig
nal
(V
)
Impulse Events
Units + EEG / LFP
Units
Multiple Neuronal (spike) Recording - two electrodes
micro-electrodes
Neurones
Amplifiernerve impulses
(action potential “spikes”)
Electrode 1
Electrode 2
Off Line Sorting of spike waveforms - single unit isolation
Movie: Sorting of unit spike data using Principal Component Analysis
• distinct AP spike waveforms represented as clusters in 3D space• 7 units isolated – each unit colour-coded
Multiple (ensemble) neurone recording
Advantages
• Investigating neuronal ensemble/network function - closer to working “brain”
• Good experimental design - fewer animals required (“3Rs” ~ Home Office)
• Masses of data
Disadvantages
• Masses of data? Data processing? Data interpretation
Data Visualisation Emergent Properties
Population Dynamics
Unit activity
Single units – spike rasters /FRH / ISIH / PSTHs burst analysis /
Unit pairs – cross-correlation / coherence /Unit ensembles – PCA / ICA / synchrony index / PDC
Local Field Potentials
FFT / spectrogramsLFP/EEG signal bandsLFP-unit coherenceLFP PDC
ENSEMBLE DATA - DISPLAY & ANALYSIS
MASSIVE data sets
• Dual/Triple site recordings ~ 64 channels simultaneous units & LFPs
• Effect of kainate (10mg/kg; i.p.) administration at epoch 4
Time Frames #1-13
corr
elat
ion
str
eng
th
PDC Analysis – unit ensemble data
• PDC was applied to identify the direction of activity between hippocampusand mPFC - technique that has the potential to reveal the neuronal ensemble drives.
Note: the magnitude of the classical coherence gives no information about directional connectivity - but its phase may do so.
University of Nottingham Medical SchoolNeuronal Networks Laboratory
University of Nottingham Medical SchoolNeuronal Networks Laboratory
Project #5:
Nociception & pain management
• dual spinal cord / supraspinal recording
Role of endocannabinoid system in normal physiology and pain (e.g. neuropathic) states
Neuronal Networks Laboratory
-2 0 2 4 6Time (sec)
10
20
30
40
sig004a
Perievent Rasters, reference = Event002, bin = 100 ms
Fre
quen
cy (
imp/
sec)
innocuous (7g) stimulation
-2 0 2 4 6Time (sec)
20
40
60
sig004a
Perievent Rasters, reference = Event002, bin = 100 ms
Fre
quen
cy (
imp/
sec)
noxious (65g) stimulation
Mechanically-evoked response in somatosensory thalamus (VPM)
• Cannabinoid receptor agonists are antinociceptive.
• CB1 predominantly expressed in the CNS but also present in the periphery.
• CB2 agonists inhibit:Acute pain [Zimmer et al.]
Inflammatory pain [Clayton et al.]
Neuropathic pain [Ibrahim et al.]
• CB2 receptors located on:Immune cellsNeuronal cells (?) [Griffin et al; Ross et al; Patel et al.]
• CB2 agonists lack CNS side effects.
Development of potent selective CB2 ligands:Agonist: JWH-133 Ki 3.4nM with a 200-fold selectivity over CB1 receptors.Antagonist: SR144528 Ki 0.67nM with a 50-fold selectivity over CB1 receptors.
Aim: To determine the involvement of the CB2 receptor in nociceptive processing.
The Role of the CB2 Receptor in Nociceptive Processing:
An in vivo electrophysiological studyNeuronal Networks Laboratory
Combined unit / LFP with USV / behavioural recording
50 ms
25
50
75
100
125
kHz
1V
296:1
296
0.05 0.10 0.15 0.20 0.25 0.30 s
25
50
75
100
125
kHz
1V
295:1 295:2295:3 295:4 295:5 295:6
295
USV call start time
unit activity - nuc accumbens
USV Recorder input
Behavioural data
Local Field Potential (LFP)
Spectrogram of specific calls [AviSoft]
Behavioural Electrophysiology
Circular arena recording using CinePlex
Movie - rat HopScotch: 8-channel array in nuc. accumbens - 6 weeks post implant
Neuronal Networks Laboratory
Recorded video
Neural data
DISCUSSION • Following KA, hippocampal units (~80%) show an increase in firing; while mPFC units show either a decrease (~80%) or increase (~20%) in firing rate. mPFC units lose their characteristic bursting pattern after KA administration.
• CCH analysis shows that unit pair activity under basal conditions is more correlated within the mPFC compared to intra-hippocampal; mPFC appears to lead hippocampal firing. KA increased correlation within the hippocampus and mPFC; but the mPFC-hippocampal drive was lost.
• PDC of unit population activity also shows basal mPFC-hippocampal directionality (predominantly at low frequencies); this initially decreases after KA administration, then later increases at all frequencies.
• In basal conditions, PDC analysis of LFPs revealed evidence of information flow from CA3 to CA1 and reciprocal hippocampal-mPFC connectivity with predominant drive from mPFC to hippocampus. Following KA, there was increased drive from mPFC to hippocampus.
This alteration in functional connectivity in a seizure model has implications for memory and learning in epilepsy.
Caveat(s):
Need to consider possible influence of anaesthesia in directing “information flow” and/or (anaesthesia/KA-induced) short-term rewiring of neural circuitry. Other regions (not recorded) may be involved in communication between mPFC and hippocampus.