Dimensionality of Rolled-up Nanomembranes Controls Neural Stem Cell Migration Mechanism

We employ glass microtube structures fabricated by rolled-up nanotechnology to infer the influence of scaffold dimensionality and cell confinement on neural stem cell (NSC) migration. Thereby, we observe a pronounced morphology change that marks a reversible mesenchymal to amoeboid migration mode transition. Space restrictions preset by the diameter of nanomembrane topography modify the cell shape toward characteristics found in living tissue. We demonstrate the importance of substrate dimensionality for the migration mode of NSCs and thereby define rolled-up nanomembranes as the ultimate tool for single-cell migration studies.


Micropatterned Trenches for Cell Culture
16 mm x 16 mm glass cover slides (0.16 mm thickness, VWR, Germany) were cleaned with acetone, isopropanol and oxygen plasma and dried for 5 min at 120 °C on a hotplate. A 20 m thick layer of negative photoresist (SU8-10, MicroChem Corp., USA) was spin-coated at 500 rpm for 10 s and 1650 rpm for 30 s onto the cleaned substrates, which were then pre-baked on a hotplate at 65 °C for 6 min and then soft-baked at 98 °C for another 10 min. The trench patterns were structured into the photoresist by gray scale lithography with a maskless aligner system (PG 501, Heidelberg Instruments, Germany). Plateau areas were exposed with 80 % lamp intensity ( = 390 ± 2 nm) for 1.5 s per pixel. 16 m wide and 6 m deep trenches were created by exposing stripes with a transversal intensity variation (stripe pattern width 20 m and 3 illumination at 8 % lamp intensity with an increase to 64 % at the rims starting 5 m away from the sides). The exposure was followed by a two-step post-baking process of 2 min at 65 °C and 12 min at 98 °C, before the samples were immersed in a bath of SU-8 Developer (MicroChem Corp., USA). Soaking in a bath of isopropanol stopped the development process. The coating of the samples with Al 2 O 3 and subsequent surface functionalization was achieved according to the same protocol as for the microtube substrate preparation. The trench profiles of the ALD-coated samples were investigated with a VK-X210 confocal microscope (Keyence Deutschland GmbH, Germany).

NSC Derivation and Maintenance
Male and female Nestin-GFP mice 4  NSCs were seeded on Al 2 O 3 -coated and fibronectin-functionalized control glass cover slides.
The cell density was adjusted to enable the observation of single cell motility either inside the microtubes or on the planar substrates one to two days after seeding. Live-cell imaging was performed using a Zeiss Axio Observer Z1 inverse microscope equipped with a 37 °C heated stage and CO 2 chamber. The software Axio Vision Rel. 4.8 (Carl Zeiss, Inc.) was used for the image acquisition. NSC motility was observed with differential interference contrast (DIC) imaging every 2 min for 4 h.

Inhibitor Experiments
Samples were prepared and imaged according to the protocol for motility experiments.

Data Analysis
The acquired fluorescent images and DIC time series were processed and analyzed with Fiji (distribution of ImageJ 1.48q, 64 bit).
For the analysis of the cell shape descriptors, Nestin-GFP fluorescence for the in vitro-grown cells or Tbr2-GFP fluorescence of the in vivo cells in fixed brain slices was recorded. For the planar glass and microtube substrates, images of the NSC circumferences were taken (Zeiss Axio Observer Z1, Zeiss HXP 120 UV lamp). For the in vivo grown cells, where the cell orientation not necessarily laid within the focal plane of the microscope, image stacks of different focal planes (z-stacks) at a slice distance of 0.5 m were taken of fixed brain slices (Zeiss LSM 780).
Afterwards, maximum intensity projections of the z-stacks were generated. A Gaussian blur filter (0.4 m scaled units) was applied to the images. They were thresholded (Fiji default settings, "dark background" active) and the cell "footprints" were characterized with the particle analyzer plugin. The cell spreading area (A), the perimeter (P) of the spreading area, the circularity (Circ) defined as: and the aspect ratio as the ratio of major to minor cell axis were evaluated.
For the quantification of the z-height and volume of the cell bodies, z-stacks at a slice distance of 0.5 m were acquired of fixed NSCs stained for actin (in vitro-grown cells) or of fixed Tbr2-GFP NSCs (brain slices). To assess the z-height of the cells, a cut view of the z-stacks was generated and the z-dimension was measured from the thresholded (Fiji default settings, "dark background" active) images. For the quantification of the cell volume, the cell area on each slice of the z-stack was measured according to the cell spreading area evaluation procedure, and then multiplied with the slice distance of 0.5 m. For the z-height and cell volume evaluation only microtubes with dimensions larger than the cell diameter were considered so that the increase in cell dimensionality inside the glass tubes was not restricted by the microtube walls. Thereby NSCs of different sizes were all included in the quantification and a bias due to a potential cell size selection through the microtube diameter was avoided.
Cell trajectories were recorded with the MTrackJ plug-in. 6
Video S2. Elongated morphology of an NSC moving inside a 10 m wide and 6 m deep trench.
Cells moving on the planar surface around the trench appear out of focus. Video S7. NSC track on a planar (2D) substrate (compare Fig. 4a).
Video S9. Appearance of blebs at both cell ends in a microtube with a diameter of 6 m.
Video S10. Inhibition of actin polymerization with a low dose of latrunculin A leads to a reduced NSC motility and spreading of the cell when moving (compare Fig. 4g).