Development and Characterization of scaffold based three-dimensional neuronal cultures

Tissue engineering, through different approaches, strives to create functional components of organs and cell types in vitro. A major bottle neck in recreating in vivo-like phenotypes in these in vitro systems is to find biocompatible materials to scaffold both as an interface with living tissues, and also to support the survival and function of the cells during in vitro culture. It is clear nowadays that 2D in vitro models poorly reflect the complexity of the body organs. This is particularly true for the brain, and many research group are working, using different approaches, to have a more appropriate system to investigate developmental processes, neuronal network dynamics and molecular pathways, as well as for drug screening. In my Ph.D., I aimed to circumvent these major bottlenecks by developing and carefully characterizing 3D neural cultures. To do so, I used optical imaging methods, particularly calcium imaging, to investigate the neuronal network dynamics of reconstructed 3D primary cultures from rats. In particular, I developed and tested new scaffolds for these culture systems. In the brain both conductivity and stiffness are important parameters that help shape neural networks. To mimick these properties of the brain, I tested the scaffolding properties of elastic organic polymers like Polydimethylsiloxane (PDMS) and gelatin. In addition, I used graphene-based materials as electrically conductive scaffolds. Through these studies, I developed a number of functional 3D-neural networks using different scaffolds with distinct structure, topology and compositions. In particular I determined that: 1) Patch substrate with a monolayer of gelatin nanofibers electrospun and crosslinked on a honeycomb microframe of poly (ethylene glycol) diacrylate (PEGDA) are suitable substrates for neuronal studies in a 3D environment. This method allows us to minimize exogenous material contact of cells and largely increase the exposure area of cells to the culture medium. Even though there are no connections in vertical dimension, I found that neurons, and especially astrocytes, have a more in vivo like morphology comparing to that on culture dish or on glass slide. We also found that neurons were preferentially located in the suspended areas of the monolayer nanofibers. Finally, calcium imaging revealed that primary neurons have a higher degree of neural activity on the patch than on glass. These results suggest that crosslinked and monolayer gelatin nanofibers closely mimic the extracellular matrix structure and allow more effective culture of primary neurons than conventional methods, thus facilitating advanced studies of neural functions as well as cell-based assays. 2) PDMS microlattices scaffolds, produced by conventional photolithography techniques, can be used as a soft scaffold for in vitro cell culture for both cell lines and primary neuronal cultures. The photomask with micro-scale dots array spin coated with photoresist is downward mounted on a rotating stage with a 45°angle to UV irradiated direction. After three irradiation times the UV exposed area could be developed to form a three dimensional (3D) porous photoresist template. PDMS is poured in and cured and a 3D PDMS lattice is obtained after etching. I determined that our 3D PDMS lattices are suitable for: a. Culturing NIH-3T3 cell line into the microlattices. We observed homogenously cell adhesion and extension to form a 3D in vitro culture. Cell nuclear shape could also be controlled by adjusting the unit-cell architecture of the lattice. b. Easy cell observation. This 3D scaffold is biocompatible and also transparent and it can be applied to study the differences between 2D and 3D cell cultures in vitro and enable cells maintain their in vivo morphology of tissue-like structures within an in vitro platform. c. Culturing primary hippocampal cells from rats. We observed a great cell-material interaction. Neurons and astrocytes grow over and among the pillars creating a 3D in vitro neuronal network. These results show that adjusting the size of 3D structure it is not only possible to control the cell growth and shape, but also to allow primary hippocampal cells to be cultured on a softer substrates and have suspended connection among the pillars reaching an unexpected three-dimensionality. 3) Graphene foam scaffold is particularly suitable for 3D-neural cultures as it permitted the formation of a modular network between cultured neurons, which is characterized by a higher firing frequency and synchronization. I was able to culture through the whole scaffolds (1-1.5 mm in heights) a viable neuronal network which, because of its third dimension, has many connections among distant neurons leading to small-world networks and their characteristic dynamics. This new in vitro model reflects some of the observation done on the nervous system dynamics in vivo. More in details, I defined: a. The presence of a Moderately Synchronized (MS) regime and a Highly Synchronized (HS) regime. The HS regime was never observed in 2D networks and was observed in the cases of highly connected portion of the network. During the MS regime, neuronal assemblies in synchrony changed with time as observed in mammalian brains. After two weeks, the degree of synchrony in 3D networks decreased, as observed in vivo. b. The absence of a disequilibrium between excitatory and inhibitory neurons by staining the culture for the GABA neurotransmitter. c. The presence of astrocytes with a ramified shape that resemble the in vivo shape that allows the engulfment of many neuronal terminals. d. The maturation of the network follow the pattern of maturation observed in vivo with a decrease synchronization and frequency. These results show that dimensionality determines properties of neuronal networks and that several features of brain dynamics are a consequence of its 3D topology. Moreover this scaffold allows to have a 3D neuronal network in vitro that has similar in vivo dynamics and could represent a better in vitro model to study some of the aspects of the nervous system. All together, these results show that to achieve complex connectivity among neurons, 3D-scaffolds with distinct electrical and mechanical properties can be applied. What makes the difference is the aim that every research group wants to achieve. In the future it could be useful to compare a 3D scaffold based in vitro model to a positive control, such as organotipic slices or acute slices, in order to understand how close we are to the network dynamics of an ex-vivo preparation. Moreover, it could be interesting to investigate the reasons behind the more elevated frequency and synchronization of the 3D neuronal networks. It could be possible that the different morphology of astrocytes affects the on-going activity of neurons or that there is a different speed of maturation of the network since we have another growing dimension (the z-plane). Beyond the basic research, it would be interesting to apply this new in vitro model to study the progression of some diseases like tumours, and in particular brain tumours, in a 3D environment and in a 3D brain culture.