Experience-dependent modifications of developing neural circuits and animal behaviours

Project Information

ZEBRAFISH PLASTICITY

Grant agreement ID: 239512

Closed project

Start date 15 June 2009

End date 14 June 2013

Funded under

• Specific programme "People" implementing the Seventh Framework Programme of the European Community for research, technological development and demonstration activities (2007 to 2013)

Total cost € 100 000,00

EU contribution € 100 000,00

Coordinated by

INSTITUT NATIONAL DE LA SANTE ET DE LA RECHERCHE MEDICALE
France

Final Report Summary - ZEBRAFISH PLASTICITY (Experience-dependent modifications of developing neural circuits and animal behaviours)

SUMMARY

One of the main goals in neuroscience is to understand how cognitive functions, such as sensory perception, are encoded by the dynamics of large neuronal networks. The main stream to study sensory perception has mainly focused on sensory stimulation and neuronal recordings of the induced neural responses. An alternative approach is the use of sensory illusions, in which sensory perception takes place in absence of physical external stimulus and, therefore, the neuronal circuit activities underlying sensory perception can be better isolated, since no components encoding sensory detection are induced and thus can not interfere with those related to perception.


PROJECT CONTEXT AND OBJECTIVES


An illusion is a perceptual misinterpretation of a real external sensory experience and may even induce perception in complete absence of any sensory stimulation.
To better understand what are the neural network mechanisms underlying sensory illusions we are using the motion after-effect (MAE), in which exposure to coherent continuous motion for a certain period of time induces motion perception in the opposite direction following the end of the moving stimulus.
Even thought this effect has been intensely study at the behavioral and single-cell physiology levels using different experimental models, the neural network representation at different sensory processing brain regions remains elusive.
For this purpose we are using the zebrafish larva as the experimental model, which allows monitoring the dynamics of large neural networks from large portion of the nervous system in a behaving organism.


One of the main goals in neurosciences is to understand how cognitive functions, such as sensory perception, are encoded by the dynamics of large neuronal networks. The main stream of perception research has mainly focused on sensory stimulation and recordings of the induced neural responses. An alternative approach is the use of sensory illusions. An illusion is a perceptual misinterpretation of a real external sensory experience inducing perception in complete absence of any sensory stimulation, therefore, the neuronal circuit activities underlying sensory perception can be better isolated from those implicated in sensory detection, since no components encoding sensory detection are induced and thus can not interfere with those related to perception.
One example of these sensory illusions is the motion after-effect (M.A.E.) in which exposure to coherent continuous motion for a certain period of time, will induce motion perception in the opposite direction following the end of the stimulus.
For this project we used the zebrafish larva as the experimental model, which with a transparent skin and small size enables monitoring the activity of large neuronal networks (representing a relevant portion of the whole brain), still with single cell resolution, in an intact, behaving vertebrate.

The main aims of this project are the following:


Aim 1. Experience-dependent short-term memory of developing large neuronal networks and motor behaviour. The motion-after-effect.
a. Behavioral kinematic effects of continous moving visual stimulation
b. The neural basis underlying the behavioural motion after-effect.

Aim 2. The motion-after-effect illusion in a zebrafish model of autism.
a. Generation of an zebrafish model of autism.
b. Behavioral kinematic effects of repetitive moving visual stimulation versus wild-type larvae
c. The neural basis underlying the enhanced behavioural motion after-effect in the zebrafish model for autism.

MAIN S&T RESULTS/FOREGROUNDS

The first goal of the research project was to build a specially designed two-photon microscope which will enable us to simultaneously monitor the dynamics of up-to 1000 neurons in an intact non-anaesthetized zebrafish larva, at the same time we present to the larva patterned or natural visual stimuli and monitor its eyes or tail motor behaviours (Figure 1a). Moreover, we have generated several zebrafish transgenic lines. Among them, a HuC:GCaMP3 line which expresses in all neurons a high-sensitive genetically-encoded calcium fluorescence sensor (Figure 1b). This line enables us to monitor the activity of virtually all brain regions with single cell resolution. Both the two-photon system and the transgenic HuC:GCaMP3 line became operational by the end of 2010.
With these techniques in hand, we found a very robust eye-rotation response to a coherent movement of the receptive field, as expected from the well described optokinetic response (OKR). Upon the presentation of a coherent motion visual stimulus covering a large portion of its field of view, the larva moves their eyes in the direction of the moving stimulus (slow pursuit eye movements) in order to stabilize the moving external world on the retina. When the eye reaches the maximal rotational angle, the larva generates a rapid movement in the opposite direction to bring the eye to its original position (saccade). See Figure 2a. Interestingly, when this stimulus was presented for a duration of at least 250 secs, following the end of stimulation, larvae generated, in absence of any sensory stimulation, eye movements in the opposite direction of the preceding visual stimulation (Figure 2a and b). This short-term memory effect lasted for 300 secs (Figure 2a and b, and Figure 3) and strongly resembles the motion-after-effect visual illusion (MAE) observed in mammals and some insects.
In correspondence to humans that perceive MAE with a slower velocity and amplitude, the zebrafish larvae show these post-stimuli eye pursuits with smaller amplitudes and at lower frequencies than those induced by the visual stimulation (Figure 2a and b).
This robust MAE behavioural effect enable us to study the neural mechanism underlying this type of short-term memory phenomenon, using two-photon Ca2+ imaging. We therefore began to monitor the dynamics of large circuits before, during and after the presentation of the conditioning coherent motion visual stimulus in intact behaving larvae.
Preliminary results identify two interesting neural ensembles within the optic tectum of the larva corresponding to direction-selective neurons. The first one are neurons that respond to the direction of the conditioning stimulus and remain silent during MAE. The second group are mostly silent during the conditioning stimulus but among them, a sub-population show synchronous calcium events at times that correspond to the eye pursuits during the MAE period (Figure 4). These spontaneous synchronous activities gradually decrease with the disappearance fo the MAE pursuits. These results suggest that within the optic tectum exist neural ensembles that correlate with visual detection and others with visual perception. The second type of neural ensembles showed the exact inverse behaviour where synchronous activities were seen in absence of visual stimulation and were suppressed during the conditioning moving stimulus.
Moreover, we have found that neurons belonging to the first group show habituation to moving stimuli in their preferred direction during the period corresponding to MAE, while neurons belonging to the second group (MAE related neurons) are not habituated, thus generating an unbalance between these two groups, which could then generate the motion illusion.
Interestingly, using transgenic larvae expressing the calcium indicator only in RGCs (ath5 promoter), we found that the conditioning stimulus does not induce habituation to any of the two direction selective RGC populations.
To further test if MAE depends on eye proprioception or eye muscular fatigue, we took advantage of optogenetics to inhibit eye movements during the conditioning stimulus. This inhibition could then be released during the MAE.
For this purpose we used transgenic larvae expressing hallorhodopsin (a chloride pump activated by light) under a pan neuronal promoter. Using a fiber optic we illuminated only the area A1, a eye movement control center. Upon illumination we were able to inhibit eye movements during the conditioning stimulus. Despite this strong inhibition, we observed robust MAE eye pursuits as for control, indeed suggesting that neither proprioception nor eye muscular fatigue play a role in the generation of MAE.
Moreover, the MAE effect is not restricted to eye movements, since a similar MAE effect was observed at the level of tail kinematics, suggesting that the motor effects reflecting the MAE are induced by sustained activities observed by monitoring tail movements
Overall, we have found that a specific group and not all directional selective neurons are required for motion perception. Since this sub-population is randomly distributed within the network, we suggest that sensory perception depends on a global activity
threshold were a minimum of number of direction-selective neurons need to be activated in order to induce motion perception.
Finally, human studies show that that autistic patients perceive an enhanced motion after-effect illusion with respect to controls. In this line, a postdoctorant in the lab has generated a mutant zebrafish line in which the MECP2 gene was specifically modified (Pietri T, et al., 2013). This gene is related to certain cases of autism and Rett syndrome. Preliminary results in the lab, show that this mutant line behaves different than wild-type showing a stronger MAE. We are therefore planning to compare the neural circuit bases of MAE between wild type and MECP2 zebrafish lines to shed line on the neuro functional modifications underlying an altered MAE in MECP2 larvae and hopefully open new ways to elucidate what are the neural circuit anomalies that may lead to autism.

Related documents

• final1-figures.pdf