Using brain stimulation methods to probe the physiology of motor

Using brain stimulation methods to probe the
physiology of motor
control from cortex to cerebellum
John Rothwell
UCL Institute of Neurology, London, UK
John Rothwell IoN
 First, some history!
John Rothwell IoN
Transcranial methods
of brain stimulation
bypass the barrier of
the scalp and skull
Nos 1, 2 and 7 are on
the motor cortex
John Rothwell IoN
Transcranial Stimulation of the Brain
 Bartholow (1874): faradic stimulation of the brain of
a patient exposed through a large ulcer on her scalp.
Movements of contralateral body.
•1900-present: neurosurgery (note sensory cortex)
•Gualtierotti and Paterson (1954) attempt repetitive transcranial
electrical stimulation of human cortex
•Merton and Morton (1980): single electrical stimulation of motor and
visual cortex.
•Barker et al (1986): transcranial magnetic stimulation
John Rothwell IoN
Experimental investigations into the functions of the human brain. By
Roberts Bartholow, M.D., Professor of Materia Medica and Therapeutics
and of Clinical Medicine in the Medical College of Ohio; Physician to the
Good Samaritan Hospital, etc
John Rothwell IoN
John Rothwell IoN
Gualtierotti & Paterson (1954): repeated stimulation at 30Hz for 40s.
Cathode right motor area, anode left motor area.
Merton (stimulated by RH Adrian) 1980:
Single high voltage pulse. Anode over
right hand area.
I will end by demonstrating
how failure of confidence and perseverance
can hold things up. I have here a device I built in
about 1947 to stimulate the brain through the scalp.
It consists of an old-fashioned gramophone motor
driving contacts which connect a condenser alternately
to a battery and then to the subject. We used
long trains of stimuli and large plate electrodes on
either side of the head. This was unsuccessful because
it became too painful before the voltage could be
turned up enough to make it effective. I now show
that using this original stimulator but with the right
sort of electrodes in the right place, and limiting the
number of stimuli to a few at high voltage, we could
have succeeded all those years ago in stimulating the
motor cortex
Merton, PA. Carmichael Lecture. J. Neurol. Neurosurg.
Psychiatry 1981;44;861-870
TMS is:
Phasic (like a peripheral nerve stimulus)
Not very focal
Does not penetrate deep into the brain
Sylvanus P Thompson
John Rothwell IoN
“Figure 8” coils are more focal at point where the two loops overlap
John Rothwell IoN
What gets activated in the cortex by TMS?
 A brief electrical stimulus recruits neural elements in the following
 Large diameter axons
 Small diameter axons
 Cell bodies (initial segment region)
 So in the cortex we might stimulate
 Axons of large diameter axons in the subcortical white matter.
 Axons of neurones in the grey matter
 Cell bodies in the grey matter
 BUT where is the current strongest? At the surface?
John Rothwell IoN
 Modelling studies suggest that:
 Currents are strong in gyral crowns
(near the skull), particularly when
the gyrus is oriented perpendicular
to the induced electric field
 This is why responses to TMS over
primary motor cortex are largest and
have lowest threshold if the coil
induces current perpendicular to the
John Rothwell IoN
Opitz et al
Individual brain: Fields induced on gyral crowns especially when coil is
oriented perpendicular to gyrus (A)
(E) Using an anisotropic model of conductivity in grey/white matter, note
that field is strong in white matter as well as in crown of gyrus. In white
matter the nerve activating function will depend on orientation of the
fibres to the field
John Rothwell IoN
 Modelling studies suggest that:
 Unexpectedly, the anisotropic conductances in the white matter and
the grey matter/white matter boundary make high induced electric
fields in the subcortical white matter.
 In the motor cortex white matter, the largest diameter axons are from
giant Betz cells that travel in the corticospinal tract; the remainder are
cortico-cortical axons
 What is stimulated depends on the orientation of the fibres w.r.t. electric
field and the relative proportions of each type of fibre….Betz cells are in a
large minority!
John Rothwell IoN
Modelling fields in precentral gyrus using anisotropic model. Note very
high field strengths in white matter.
Fibres that are most likely to be activated by this are those that bend into
the white matter from the grey matter.
John Rothwell IoN
 Sites of activation are therefore in grey matter at gyral crown and subcortical
white matter. In the latter the fibres activated preferentially will tend to bend
into/out of the direction of the induced electric field
 In motor cortex this causes:
 Lowest threshold effects are inhibitory (is this grey matter??)
 Next lowest threshold is I-wave activation (is this white matter??) with the precise
set of inputs depending on the direction of induced current flow (I1 inputs with
posterior-anterior induced current; I3 inputs with anterior-posterior induced current)
 Higher threshold are axons of corticospinal neurones in White matter (D-inputs)
(particularly if current is in latero-medial direction
 Transcranial high voltage electrical stimulation leads preferentially to D-activation
John Rothwell IoN
Currents induced by standard “Magstim” 200 stimulator and coil:
NOTE threshold is different for each direction, lowest with PA,
highest LM
PA induced
(I1 waves)
AP induced
(I3 waves)
LM induced
(D waves)
Each direction preferentially activates different populations of cortical
neurones. They have different effects on the brain!
John Rothwell IoN
Anodal stimulation
volleys from
space and
EMGs in
Magnetic stimulation
during active
contraction to
show shortest
20 uV
5 ms
1 mV
10 ms
John Rothwell IoN
Short interval intracortical inhibition (SICI)
John Rothwell IoN
I1 I3
ISI 1ms
ICI showing
volleys in
space and
ISI 2ms
ISI 3ms
ISI 4ms
ISI 5ms
shock alone
10 µV
5 ms
1 mV
5 ms
John Rothwell IoN
TMS: inhibition after excitation
 A single stimulus will excite neurones
synchronously. In the cortex this often
produces a short burst of rapid firing
of cells, followed by a longer period of
inhibition, rather like a “spike-wave” in
 This produces the “silent period” that
follows the MEP (GABAb)
 The net result of all this is that any
processing in that area that is going on
at the time the stimulus is given will
be disrupted
 Basis for “virtual lesion” studies
John Rothwell IoN
No of correct letters
Transient scotoma (“virtual lesion”) produced
by stimulation over visual cortex (Amassian et al)
Very brief presentation
of three dim letters on
screen. Subjects identify
the letters.
subject 1
subject 3
Time of TMS after visual presentation
Give TMS to occiput
after letters flashed
At correct timing
subjects can no longer
see anything
John Rothwell IoN
 TMS to motor cortex activates many outputs in addition to those in
corticospinal tract that produce MEPs
 These other projections can be visualised with other methods such as
John Rothwell IoN
John Rothwell IoN
Brain activations obtained for a group
analysis (9 subjects, P < 0.01,
corrected) of responses to
suprathreshold rTMS of the left PMd. (a)
Sagittal (x=-40), coronal (y=-11), and
transverse (z=55) view of activity in the
left PMd. (b) Six transverse sections
showing activity changes in the CMA,
PMv, auditory cortex, caudate nucleus,
left posterior temporal lobe, medial
geniculate nucleus, and cerebellum.
Activation maps are projected on a
template brain (Montreal Neurological
Institute, MNI)
John Rothwell IoN
Spread of activation from TMS
stimulus over PMd
Ferarrelli et al (2010)
blue traces for waking, red traces
for midazolam anaesthesia
“…might be possible to use TMS-EEG to
assess consciousness during anesthesia
and in pathological conditions, such as
coma, vegetative state, and minimally
John Rothwell IoN
conscious state”
Recruitment of alpha EEG
activity by 5 TMS pulses
given at alpha frequency(10
Note how the response to
the initial 2 pulses is
widespread, but at the last
3, is focussed at the site of
stimulation, and gradually
increases in magnitude
(Thut et al., 2011)
John Rothwell IoN
After-effects of prolonged stimulation: synaptic plasticity
 Occur after repetitive TMS protocols or after several minutes of
continuous TDCS (or static magnet application)
 Detected in motor (and visual) cortex by lasting (30min typical) effects
on motor or visual thresholds as tested by single pulse TMS (MEPs,
 Effects abolished by drugs that interfere with NMDA receptors
 Thought to represent the early stages of synaptic plasticity leading to
LTP/LTD at cortical synapses
 Can interact with behavioural learning
John Rothwell IoN
Strengthening synaptic
Record here
Repeated activation of an
existing synapse can increase
its effectiveness
Long term potentiation (LTP)
Short period
of repetitive
stimulation here
Evidence that an rTMS paradigm (iTBS) may produce after effects
on cortex due to plasticity at cortical synapses (Huang et al, 2009)
Effects of theta burst stimulation on motor cortex are blocked
by memantine, an NMDA receptor antagonist
Transcranial Direct Current Stimulation (TDCS)
Apply 1-2mA through
large scalp electrodes
for 5-20 min
Polarises neurones in
cortex. No action
Anodal TDCS tends to
depolarise, cathodal to
John Rothwell IoN
TDCS Physiology
 Experiments on cat and rat cortex showed that DC
polarisation of the cortical surface changed the firing rate
of pyramidal neurones
 Anodal (+) stimulation increased firing rates
 Cathodal (-) stimulation decreased firing rates
 Currents are of the order of 2.5 mA/cm2
 Bindmann et al then found that the effects outlasted the
DC stimulation by many minutes and hours
 After-effects could build up over 30min or so when
stimulation terminated
John Rothwell IoN
Cathodal (outward)
Creutzfeld, Fromm & Kapp
(1962). Cat cortex
Anodal (inward)
John Rothwell IoN
Surface anode
will depolarise
deep layers
Note that although
Creutzfeld et al found
anodal usually increased
firing rates of motor
cortical neurones, it
sometimes decreased
rates in neurones
recorded from the
bottom of a sulcus.
Perhaps some of them
are oriented in opposite
direction w.r.t. the
John Rothwell IoN
After-effects can last hours
But note this is with
polarisation through recording
electrode in grey matter
John Rothwell IoN
TDCS in mouse
M1 increases
evoked LFPs and
is blocked by
NMDA receptor
Fritsch et al,
John Rothwell IoN
TDCS: mechanisms of long lasting effects?
 Induction of 50 Hz LTP can be increased if depolarise during
 Spike timing dependent plasticity can be increased by
depolarising the post-synaptic cell
John Rothwell IoN
TDCS: humans
 Very similar to the animal data:
 Concurrent anodal TDCS increases MEPs evoked by TMS
(presumably due to depolarisation of neurones by TDCS)
 After-effects of TDCS seen as changes in MEP amplitude in 30min
following TDCS
 BUT much lower currents
 But effects in human are very weak, only detectable with TMS.
 Also in human the effects are very variable.
John Rothwell IoN
Realistic head modelling of currents in brain
Surface inflation – note fields at depths of sulci
Cortical interface
Normal component of E-field
A negative normal component means that current is flowing into the cortex.
Static magnets (10min) suppress MEPs
for 20min
Large Magnet strength: 76kg
Small magnet: 23kg
John Rothwell IoN
Tufail et al (2010)
Pulsed ultrasound
stimulation of deep
brain structures
(ultrasound can be
John Rothwell IoN

similar documents