Importantly, the size features of the ridges on the line substrate are smaller than a growth cone. Furthermore, we observed that the neurite is slightly deflected compared to the ridge direction. Orientation of neurite outgrowth does therefore not happen by physical trapping of the neurite in the grooves. Thus, the simple fact of altering the topographical state of which an ECM is presented to the cell drastically alters neurite orientation and outgrowth. Neurite orientation not only occurred with our neuronal-like neuroblastoma cell line, but similar results were also observed with freshly isolated primary cortical neurons that were plated on a 1:5 line substrate coated with poly-L-ornithine and laminin. We next thought to understand the cellular mechanisms that allow the specific neuronal cell responses on the line substrate. For that purpose, we used the 1:5 line substrate throughout this study since it leads to the most robust phenotype in terms of neurite length. We first immunostained the cells on plain and 1:5 line substrates to visualize the F-actin and tubulin cytoskeletons 2 and 24 hours after plating. Surprisingly, we found that a higher amount of filopodia was typically observed on the soma, neurite shaft and growth cone of cells on plain versus line substrate. Quantitation LY294002 revealed a two fold increase of filopodia number on the neurite shaft on plain versus line substrate. These filopodia were also longer. While growth cones were highly spread and displayed a high density of randomly oriented filopodia on plain substrate, less spread, streamlined growth cones with fewer filopodia occurred on line substrate. These growth cones exhibited thick filopodia that aligned in the direction of the pattern ridges and displayed a high F-actin content as observed by phalloidin staining. This was especially evident with high resolution images of growth cones on the line substrate, and, in addition to the thick, F-actin rich aligned filopodia revealed a second population of thin, F-actin poor filopodia that were not aligned with the lines. Similar results were also observed in SEM experiments and revealed that thick filopodia align and intimately adhere along the top of the line ridges, whereas thin, unaligned filopodia only interact with the line ridges at discrete points. We then used phase contrast time-lapse microscopy to study the Adriamycin morphodynamics of neurite outgrowth on plain and line substrates. We observed that neurites exhibited a highly unstable behavior that consisted of multiple cycles of neurite protrusion and retraction events on the plain substrate.