In this work, we analyze the impact of a chip coating with a self-assembled monolayer (SAM) of (3-aminopropyl)­triethoxysilane (APTES) on the electronic and mechanical properties of neuroelectronic interfaces. We show that the large signal transfer, which has been observed for these interfaces, is most likely a consequence of the strong mechanical coupling between cells and substrate. On the one hand, we demonstrate that the impedance of the interface between Pt electrodes and an electrolyte is slightly reduced by the APTES SAM. However, this reduction of approximately 13% is definitely not sufficient to explain the large signal transfer of APTES coated electrodes demonstrated previously. On the other hand, the APTES coating leads to a stronger mechanical clamping of the cells, which is visible in microscopic images of the cell development of APTES-coated substrates. This stronger mechanical interaction is most likely caused by the positively charged amino functional group of the APTES SAM. It seems to lead to a smaller cleft between substrate and cells and, thus, to reduced losses of the cell’s action potential signal at the electrode. The disadvantage of this tight binding of the cells to the rigid, planar substrate seems to be the short lifetime of the cells. In our case the density of living cells starts to decrease together with the visual deformation of the cells typically at DIV 9. Solutions for this problem might be the use of soft substrates and/or the replacement of the short APTES molecules with larger molecules or molecular multilayers.

In this paper, we demonstrate that cell adhesion and neuron maturation can be guided by patterned oxide surfaces functionalized with organic molecular layers. It is shown that the difference in the surface potential of various oxides (SiO2, Ta2O5, TiO2, and Al2O3) can be increased by functionalization with a silane, (3-aminopropyl)-triethoxysilane (APTES), which is deposited from the gas phase on the oxide. Furthermore, it seems that only physisorbed layers (no chemical binding) can be achieved for some oxides (Ta2O5 and TiO2), whereas self-assembled monolayers (SAM) form on other oxides (SiO2 and Al2O3). This does not only alter the surface potential but also affects the neuronal cell growth. The already high cell density on SiO2 is increased further by the chemically bound APTES SAM, whereas the already low cell density on Ta2O5 is even further reduced by the physisorbed APTES layer. As a result, the cell density is ∼8 times greater on SiO2 compared to Ta2O5, both coated with APTES. Furthermore, neurons form the typical networks on SiO2, whereas they tend to cluster to form neurospheres on Ta2O5. Using lithographically patterned Ta2O5 layers on SiO2 substrates functionalized with APTES, the guided growth can be transferred to complex patterns. Cell cultures and molecular layers can easily be removed, and the cell experiment can be repeated after functionalization of the patterned oxide surface with APTES. Thus, the combination of APTES-functionalized patterned oxides might offer a promising way of achieving guided neuronal growth on robust and reusable substrates.

No abstract