Recapitulating neural tube morphogenesis with human pluripotent stem cells

Neural tube morphogenesis is the �rst step in the formation of the nervous system, and tube defects are among the highest rate birth defects. However, it is not possible to study the dynamics of organ formation in humans. Animals differ from humans in key aspects, and in particular in the development of the nervous system. Conventional organoids are neither reproducible nor do they recapitulate the intricate anatomy of the neural tube. Here we describe a new protocol for recapitulating the intricate dynamics of neural tube morphogenesis using human stem cells. Our approach is reproducible, scalable, compatible with live imaging, genetic modi�cations, and drug screening. The protocol captures the early steps of neural development: pattering of the ectoderm into a neural plate and surface ectoderm, neural plate folding and closure, coverage by surface ectoderm, and neural crest cells differentiation and migration. This protocol opens up a way for further studies into the genetics and biophysics of the development of the human nervous system in health and disease.


Introduction
Recent advances in 3D stem cell cultures enable studying the genetic and cellular processes which underlie the development of the human nervous system [1][2][3][4][5] .The key to these protocols is the aggregation of human pluripotent stem cells (hPSC) into three-dimensional embryoid bodies (EBs), followed by embedding in Matrigel to drive their expansion.This approach enables differentiation into diverse cell types, and self-organization in 3D which are reminiscent of brain development.However, this approach suffers from large variability between samples, and does not accurately model the 3D organization of the early nervous system -i.e. a neural tissue encapsulating a single large lumen 6 .Here, we present a protocol which solves these issues, and enables us to recapitulate the early steps of neural tube development.

Procedure
Our approach is to recreate as closely as possible the initial and boundary conditions, physical and biochemical, in the embryo at the onset of neural tube development.The nervous system develops from the anterior epiblast, a pluripotent epithelium overlaying the endoderm, and is in contact with a lumen -the amnion.Neural development is triggered by a combination of NODAL and BMP signaling from surrounding tissues 7,8 .Inspired by these events, we used micropatterning of stem cells [9][10][11] to create a pluripotent epithelium with controlled and reproducible dimensions (Fig. 1, Day 1).We than used low concentration Matrigel to drive robust folding into a 3D epithelium containing a single lumen.(Fig. 1, Day 3).Finally, we exposed the culture to TGFβ inhibition and BMP.The precisely controlled initial and boundary conditions of the sample triggered robust pattern formation and folding morphogenesis (Fig. 1, Day 7).In the following sections we describe in details the steps of the protocol.Key steps include microfabrication of PDMS stamps (4hrs), micropatterning of glass-bottom dishes (2hrs), stem-cell seeding (2hrs), lumen formation and differentiation protocol (9 days).Finally, we describe catheterization of neural tube morphogenesis using whole-mount immunostaining.
Step 1: Microfabrication of PDMS stamps This step explains how to create the PDMS stencils, which are required for micro-patterning of stem cells.
The fabrication of SU8 molds is described, followed by PDMS casting and curing.These steps have been previously described in detail 12 and are brought here for completeness.The stencil fabrication and micropatterning steps can be skipped as micropatterned dishes are commercially available and successfully applied 9 .However, we nd that in-house manufacturing is low-cost, fast, and offers experimental exibility.Wafer Prep 1. Nano-strip at ~70°C for at least 5 minutes.2. Degas air bubbles in vacuum.
4. The PDMS layer is then peeled off the silicon mold and individual stamps are cut out using a razor blade for future use.
Step 2: Micro-contact printing This step describes how to create extracellular matrix (ECM) protein micropatterns.In previous publications 12 , a PDMS stencil is incubated in a bath of ECM protein, and then brought into contact with a glass surface for a short period of time.Then the microprinted surface is coated with PLL-PEG to passivate the glass surface outside the ECM printed regions.Our approach is slightly modi ed.We rst apply PLL-PEG to passivate the surface, while protecting regions of interest using a PDMS stencil (Fig. 2).Then in a second step we expose the entire surface to ECM proteins and those assemble in the areas previously protected by the PDMS stencil.In our hands this approach has been more robust than contactprinting of the ECM protein.
1. Drill 20mm hole in the bottom of 35mm petri dish.
2. Clean glass coverslip and 35mm dish using ethanol.
3. Apply a small amount of NOA-81 to dish bottom.Place coverslip on dish and cure using UV light.
5. Prepare PDMS stamp, 1cm x 1cm in area by sonicating stamps in ethanol for 10min and drying with an airgun.6. Place cleaned dishes and PDMS stamps in plasma oven.Pull a vacuum for 30-60s.Using needle valve allow a small stream of ambient air into the chamber.Turn on radiofrequency power to full (18W) to generate a purple-hued plasma for 60s.Turn off the radiofrequency power.Turn off pump.Vent slowly.
7. Place PDMS stamps in contact with glass.
8. Place R621 magnet on PDMS top, R821 magnet on bottom-side of glass coverslip.9. Use 200ul pipet tip to add PLL-PEG solution around PDMS stamps.The PLL-PEG will ow between the PDMS and glass coverslip by capillary forces.Then cover entire glass and plastics (Total 200µl per dish).10.Incubate for 30min.
12. Carefully remove PDMS stamp and wash with PBS++ two times.13.Coat with a solution of Laminin-521 (5% v/v in PBS++) and incubate overnight at 4C (750µl Laminin solution per dish).
14. Wash micropatterns with PBS++.Micropatterns are now ready to use.
15. Quality control.To verify that micropatterns have formed, apply a uorescently tagged protein (e.g.SA-488, Thermo sher S11223) for 10min.Wash and observe in a uorescence microscope.
Step 3: Stem-cell seeding (Day 1) This step describes seeding hPSCs onto the ECM micropatterns.The protocol is adapted from Warm ash et al. 13 , and there are slight modi cations to incubation times and washing steps.

Add 200μl
ReLeSR per well and incubate for 2min at RT. 4. Aspirate ReLeSR and incubate for 5min at 37C. 5. Resuspend in 1mL mTESR per well.6. Centrifuge cells at 180RCF for 3min into a pellet.7. Aspirate media, and resuspend cell pellet into mTESR supplemented with 10μM ROCK inhibitor.Final volume of resuspension should be 400μl per well.
8. Pipet cell solution to form a single cell suspension using 1ml pipette tip.Pipet up-down 10 times.9. Pipet 200μL of cell suspension into the center of prepatterned dishes.Allow cell suspension to cover the glass area by gently tilting the dish.Avoid liquid from reaching the non-patterned plastic area as this will reduce cell surface density.10.Incubate 15min at 37C in an incubator.
11. Add 1mL of mTeSR1 per well and incubate for 10 min at 37C incubator.
12. Excess media is aspirated, leaving enough liquid to cover patterns and replaced with fresh 2mL of mTeSR1.
13. Check micropatterns in binocular.Cells should be con ned to the micropatterns, with some cells left between patterns.Don't attempt to remove these cells, as the culture has not adhered strongly at this step.Cells will detach spontaneously from these regions overnight.
Step Lumen formation (Days 2-3).This step describes how to transform 2D micropatterned hPSC colonies into 3D colonies with a single lumen.For this purpose we apply Matrigel, which has been previously used to generate lumens from small hPSC colonies 14 and in brain organoids 2 .We nd that exposure of 2D hPSC cultures to low percentage Matrigel results in robust lumen formation.This occurs through folding of the 2D hPSC culture, and maintains the epithelial structure of the culture.Thus, the initial epithelial state of the culture is essential for the formation of a single lumen.This is in contrast to Matrigel embedding of brain organoids 2 in which the cell aggregate lacks epithelial organization prior to Matrigel exposure, and exposure to Matrigel results in the formation of multiple lumens.Another key point is the media in which Matrigel is diluted.Here we describe addition of Matrigel diluted in neural induction media, which is optimized for the neural tube morphogenesis protocol.However, to maintain the pluripotency of the culture during the transition to 3D, one should dilute Matrigel in mTESR.This allows further differentiation into mesendoderm cell types.Here we describe the steps for triggering neural induction and folding morphogenesis in 3D stem-cell cultures.The approach is based on exposure to BMP4, which is known to pattern the ectoderm into neural and non-neural domains in a concentration dependent manner 8 .A similar approach has been used in 2D stem cell cultures to drive pattern formation, however folding morphogenesis was not observed [9][10][11] .Thus, the presence of a 3D stem cell culture, and a lumen, are essential for folding morphogenesis.The timing of BMP addition is critical, and we nd that a minimal neural induction period of three days is required before adding BMP.Adding BMP earlier than that results in mesendoderm fates and does not lead to folding morphogenesis.

1 .
Media change (Day 4).a. Check for lumen formation in bright eld.Patterns should be circular though maybe somewhat retracted from original micropattern.A single lumen should appear in each colony wrapped by a smooth epithelial layer (Fig. 1b, day 4).b.Change N2 media with fresh N2 media supplemented with 5μM of TGFβ-inhibitor SB-431542.

Figures
Figures

Figure 1 Key
Figure 1

Figure 2 Steps
Figure 2