The three main phases of neural development

1
(No Transcript)
2
The three main phases of neural development
1. Genesis of neurons (and migration).
2. Outgrowth of axons and dendrites, and
synaptogenesis.
3. Refinement of synaptic connections.
3
Ocular dominance and monocular deprivation
This led to the competitive interactions
hypothesis, which can explain circuit refinement
4
Explanations of OD plasticity
Hebbian plasticity

• When the pre-synaptic and the post-synaptic
  neurons fire together their synapse is
  strengthened (LTP).
• When the pre-synaptic and the post-synaptic
  neurons do not fire together their synapse is
  weakened (LTD).
• Thus, in binocular neurons, the synapses from
  the closed eye are weakened, while the synapses
  from the open eye are potentiated.

Homeostatic plasticity

• This concept is founded on the observation that
  neurons can maintain their responsiveness (e.g.
  firing rate) within a preferred range in spite of
  chronic alterations of neuronal activity levels.
• - Thus, visual responsiveness of deprived neurons
  could be enhanced directly, without Hebbian
  plasticity, by increasing synaptic strength,
  intrinsic excitability, or both.

5
2-photon calcium imaging (opto-physiology)
Green- neuron somata Red- glia
- Used for non-invasive measurement of optical
activity of dozens of cells, with single-cell
resolution.
- Compared with multi-electrode recording,
optical population recording has the advantage
that all of the cells in a field of view can be
probed, regardless of whether or not they are
firing action potentials.
- As their spatial locations are precisely known,
the cell types of all recorded cells can be
determined.
- Calcium signal spiking activity
6
(No Transcript)
7
Mapping Ocular Dominance (OD) in Mouse Binocular
Visual Cortex by Two-Photon Calcium Imaging (1)
8
Mapping Ocular Dominance (OD) in Mouse Binocular
Visual Cortex by Two-Photon Calcium Imaging (2)
9
So we can look directly at OD with single-cell
resolution, now lets examine plasticity
10
OD Plasticity
- Just like in the classical OD experiments,
contra eye monocular deprivation (MD) shifts the
distribution to the ipsi eye, and vice versa.

• Generally
• The response to the deprived eye is decreased.
• The response to the non-deprived eye is
  increased.

• Timescales
• After 1 day of MD there is no change.
• After 2-3 days the main effect is the reduction
  of deprived eye responses.
• Depression precedes potentiation during OD
  plasticity in rodents (as was shown previously).

11
What are the possible mechanisms of OD plasticity?
- Deprived-eye response depression can be
explained by homosynaptic LTD of excitatory
synapses or by LTP of inhibitory synapses
(Hebbian plasticity).
- The increased visual drive from the
non-deprived eye could be mediated equally well
by LTP or non-Hebbian, compensatory mechanisms
(homeostatic plasticity).
- If the increased visual drive from the
non-deprived eye is mediated by homeostatic
plasticity, what can we predict to test this
experimentally?
12
Predictions for testing the contribution of
homeostatic plasticity

• The proportion of neurons responding
  preferentially to the deprived eye (monocular
  neurons) should not change after MD (as the LTD
  is compensated to retain homeostasis).
• The responses of these monocular neurons should
  be increased after eye reopening, as they should
  have increased their responsiveness to the
  reduced visual drive through the closed eyelid.
• The duration of MD necessary for an increase in
  deprived-eye response in these monocular neurons
  should match that required for delayed
  strengthening of open-eye responses in binocular
  neurons (as they presumably arise from the same
  homeostatic mechanism).
• If homeostatic mechanisms act to maintain
  neuronal firing rates within a certain range, the
  strength of eye-specific inputs should be
  adjusted such that the combined visual drive from
  the two eyes remains roughly constant.

13
Testing the contribution of homeostatic
mechanisms to OD plasticity (1)
- The proportion of monocular, deprived-eye
neurons, in deprived animals was no different to
the proportion of these neurons in controls
(supporting prediction a).
- The entire deprived-eye response range of
neurons responding predominantly or exclusively
to the deprived eye (OD score 00.25) was shifted
to higher response values after contralateral-eye
MD (supporting prediction b).
(This observation is best explained by
homeostatic mechanisms acting independently of
Hebbian learning rules).
14
Testing the contribution of homeostatic
mechanisms to OD plasticity (2)
- The decrease in the response to the closed eye
in binocular neurons was full after 2-3 days of
MD.
- The increase in the response to the closed eye
in monocular neurons was only full after 4-7 days
of MD, just like the general increase in
binocular neurons (supporting prediction c).
monocular
This further validates the existence of separate
mechanisms for (fast) Hebbian plasticity and
(slow) homeostatic plasticity.
15
How can we verify that these neurons indeed
received the majority of their inputs from the
deprived eye before MD?
- One approach to verify this is to perform
chronic recordings from the same animal before
and after MD, but this is technically difficult.
- Another reasonable possibility is to measure
visually evoked responses in the monocular
cortex of normal and monocularly deprived mice
and see if the same effect is evident.
- This effect (and the calcium imaging as a
measure of spiking activity) was also verified
with electrophysiological recordings.
16
Is this effect also evident with binocular
deprivation (BD)?
- The increased response was evident, as
expected, also in BD (5-6 days).
This result confirms that response depression in
binocular cortex occurs only during MD, when the
activity through one eye is higher than in the
other. If not then there is a general
potentiation.
17
Back to testing the contribution of homeostatic
mechanisms to OD plasticity (3)
Reminder (prediction d) If homeostatic
mechanisms act to maintain neuronal firing rates
within a certain range, the strength of
eye-specific inputs should be adjusted such that
the combined visual drive from the two eyes
remains roughly constant.
18
Summary and conclusions (1)

• In neurons with significant open-eye input,
  deprived-eye responses were reduced while those
  of the open eye increased.
• In contrast, the deprived-eye responses of
  neurons largely devoid of open-eye input were
  stronger after MD.
• Therefore, the direction of the shift of
  deprived-eye responses in each cell depended
  critically on the amount of open-eye input and
  net visual drive experienced during MD.
• Consistent with these findings, responses to both
  eyes were up-regulated after BD.

19
Summary and conclusions (2)

• The proportion of neurons responding
  preferentially to the deprived eye did not change
  after MD (as the LTD is compensated to retain
  homeostasis).
• The responses of these monocular neurons were
  increased after eye reopening, as they have
  increased their responsiveness to the reduced
  visual drive through the closed eyelid.
• The duration of MD necessary for an increase in
  deprived-eye response in these monocular neurons
  did match that required for delayed strengthening
  of open-eye responses in binocular neurons (as
  they presumably arise from the same homeostatic
  mechanism).
• Homeostatic mechanisms probably act to maintain
  neuronal firing rates within a certain range as
  the strength of eye-specific inputs were adjusted
  such that the combined visual drive from the two
  eyes remains roughly constant.

20
The proposed sequence of events in
binocular neurons (a combination of Hebbian and
homeostatic plasticity)

• The decorrelated input though the closed eye
  initially causes a Hebbian weakening (LTD) of
  deprived-eye synapses during the first few days
  of MD.
• Subsequently, this triggers a compensatory
  (homeostatic) upscaling of responses to the
  open eye.
• This keeps the total post-synaptic drive constant
  (in homeostasis).

21
Question?