Digital archive of theses discussed at the University of Pisa


Thesis etd-10022015-130232

Thesis type
Tesi di laurea magistrale
Thesis title
Neuroplasticity in the adult visual cortex: role of Semaphorin 3A aggregation by perineuronal nets and development of a custom apparatus for longitudinal evaluation of visual cortical plasticity using Intrinsic Signal Optical Imaging
Course of study
relatore Prof. Pizzorusso, Tommaso
tutor Prof. Vignali, Robert
  • perineural nets
  • neuroscience
  • neuroplasticity
  • intrinsic signal
  • imaging
  • plasticity
  • semaphorin
Graduation session start date
Release date
During development, the brain goes through defined temporal windows of enhanced plasticity called Critical Periods. One of the hallmarks of the closure of these phases is the activity-dependent organization of the extra-cellular matrix (ECM) in densely organized structures called peri-neuronal nets (PNNs). The PNNs are composed of extracellular matrix molecules, including hyaluronic acid, link proteins (e.g., Crtl1) and chondroitin sulfate proteoglycans (CSPGs) and they shroud preferentially cortical parvalbumin positive (PV+) fast spiking interneurons, a subclass of inhibitory GABAergic neurons.
The causal relationship between the organization of the ECM in these structures and the restriction of plasticity is emerging as very clear from a wide set of experimental evidences in both mice and rats. For example, mice lacking Crtl1, a key component for PNN stability, retain cortical plasticity into adulthood; furthermore, strikingly, the enzymatic digestion of CSPGs though intracortical injections of Chondroitinase ABC reactivates plasticity even in adult rats.
Despite this growing set of data addressing the role of PNNs in neuroplasticity, it is still unclear which are the precise molecular mechanisms mediating their restrictive role towards plasticity.
Semaphorin 3A (Sema3A) is a secreted protein that exhibits a chemorepulsive role for axonal growth. Recently it has been showed that Sema3A has a high biochemical affinity towards different classes of CSPGs (e.g., aggrecan, versican) and in histological preparations, it is strongly enriched around PNN-positive neurons. Intriguingly, both in mice lacking the link protein Crtl1 and in rats intracortically injected with Chondroitinase ABC, Sema3A does not concentrate around neurons and remains diffuse, suggesting that the inhibitory function of PNNs could be due to their competence in binding and accumulating repulsive soluble proteins present in ECM and presenting them in high concentrations to the neurons they enfold.
In order to address this hypothesis, we interfered with the extracellular activity of Sema3A through the injection, in the visual cortex of adult rats, of adeno-associated viruses (AAV) that drive the expression of the “receptor body” neuropilin1-Fc. This protein contains a binding domain for Sema3A, but is soluble and present in the extracellular space, thus being unable to activate the intracellular signaling pathway that Sema3a would initiate. As a result, the receptor body competes with the transmembrane receptor for the binding of Sema3A and strongly reduces the relevance of its signaling pathway thanks to the strong concentration imbalance favoring the soluble and inactive form. After 3 weeks from the injection we tested whether the simple inhibition of Sema3A reactivates plasticity in adult rats by using the classical experimental paradigm of Monocular Deprivation (MD) to elicit plastic phenomena in the visual cortex.
Normally, the binocular part of the rat’s visual cortex receives inputs from both eyes, but is driven more strongly by the contralateral eye, a property called Ocular Dominance (OD; i.e., the dominance of the contralateral inputs on the ipsilateral ones). By perturbing the cortex through the suture of the contralateral eye, and thus reducing the amount of information flowing through the dominant pathway, a plastic process called Ocular Dominance Shift can be elicited. The cortex responds to the manipulation favoring the inputs from the open, ipsilateral eye and/or progressively ignoring the silent contralateral eye. This phenomena is possible if and only if the cortex is still plastic (e.g., in young animals, but not in adult ones) and thus can be used to precisely assess the presence of plasticity in the visual cortex.
After 7 days of monocular deprivation, we used in-vivo electrophysiology to measure the amplitude of visually evoked potentials (VEPs), low frequencies potentials that are generated in the visual cortex in response of a stimulus. By comparing the amplitude of VEPs generated by the stimulation of the ipsilateral and the contralateral eye, we showed that the inhibition of Sema3A pathway reactivates ocular dominance plasticity in the adult rat visual cortex.
To better understand the relationship between Sema3A and PNNs, we examined their localization by immunohistochemical (IHC) staining in slices of the visual cortex of dark reared rats. Dark rearing is a condition known to dramatically delay the formation of PNNs and to prolong the Critical Period for Ocular Dominance Plasticity. We found that Sema3A positive cells are significantly rarer in dark reared animals, supporting the hypothesis that PNNs are essential to bind and accumulate soluble molecules to create local spots of high concentration.
Even if interfering with the function of Semaphorin-3A can reactivate Ocular Dominance Plasticity in adult animals, it is still unknown whether the shift in Ocular dominance observed following MD is due to a decrease of the responses from the contralateral eye, an increase of the responses from the ipsilateral eye, or both. The importance of these data resides in the fact that these two types of plasticity are well studied in young animals and are known to be based on different physiological mechanisms. Understanding how Sema3A regulates neuroplasticity is therefore important to comprehend whether the disruption of its signaling pathway engages one, the other, or both types of plastic changes. Unfortunately, this kind of data is quite hard to obtain because it requires longitudinal measures of cortical responses in the same animal, while in our lab, VEPs recording is mainly an acute procedure thus yielding only cross-sectional measures.
In order to be able to address this important question, we developed both the hardware and the software of a custom apparatus for Intrinsic Signal Optical Imaging, and we performed several steps of testing, with the goal of being able to record cortical evoked responses chronically and with extremely low invasiveness.
“Intrinsic signals” are optical signals that can be used to measure neural cortical activity. They are characterized by a relatively slow (i.e., with a time constant in the order of seconds) decrease in light reflectance that is more evident at 630nm. They are generated in response of neural activity by a combination of a local increase in blood volume due to neurovascular coupling and by an increase in the proportion of deoxygenated Hemoglobin, which has slightly different optical properties from the oxygenated form. Importantly, given their physiological nature, these signals are detectable in naïve animals without using any external dye or genetically encoded indicator, so they are great candidate for low-invasiveness studies.
The hardware for imaging is composed by a standard optical microscope, a low-noise camera (CMOS sensor), and a custom imaging chamber to accommodate the animal during the procedure. The imaging chamber has been designed and 3D printed specifically to accommodate two automated eye shutters, and a heating pad, all remotely controlled by an Arduino microcontroller.
The software has been written in MATLAB and comprises scripts to control a “stimulator” computer that generates the visual stimuli and a “recording” computer that acquires images from the camera. In addition, we developed specific algorithm and Graphical User Interfaces (GUIs) for further data processing. Specifically, we implemented algorithms for image visualization and filtering, timeline visualization, movie visualization, artifact rejection and comparison of ipsilateral and contralateral recordings.
Our goal was to use this setup for chronic imaging in adult animals to understand the physiological effects of interfering with the ECM, with the least possible amount of invasiveness, potentially through the intact skull. We performed several steps of testing to validate our apparatus: first, we addressed the possibility to record through the intact skull in young mice; second, we were able to detect the shift in Ocular Dominance produced by MD in young mice; third, we showed the feasibility of chronic recordings through the intact skull in adult mice, a challenging procedure, given the thickness and opacity of the adult skull. In the future, we plan to extend this technique also in rats to have a wider spectrum of possibility of investigation, and to better elucidate how the ECM regulates plasticity.