High-Content Screening Differentiation and Maturation Analysis of Fetal and Adult Neural Stem Cell-Derived Oligodendrocyte Precursor Cell Cultures

Summary

We describe the production of mixed cultures of astrocytes and oligodendrocyte precursor cells derived from fetal or adult neural stem cells differentiating into mature oligodendrocytes, and in vitro modeling of noxious stimuli. The coupling with a cell-based high-content screening technique builds a reliable and robust drug screening system.

Abstract

The main hurdle in developing drug screening techniques for assessing the efficacy of therapeutic strategies in complex diseases is striking a balance between in vitro simplification and recreating the complex in vivo environment, along with the main aim, shared by all screening strategies, of obtaining robust and reliable data, highly predictive for in vivo translation.

In the field of demyelinating diseases, the majority of drug screening strategies are based on immortalized cell lines or pure cultures of isolated primary oligodendrocyte precursor cells (OPCs) from newborn animals, leading to strong biases due to the lack of age-related differences and of any real pathological condition or complexity.

Here we show the setup of an in vitro system aimed at modeling the physiological differentiation/maturation of neural stem cell (NSC)-derived OPCs, easily manipulated to mimic pathological conditions typical of demyelinating diseases. Moreover, the method includes isolation from fetal and adult brains, giving a system which dynamically differentiates from OPCs to mature oligodendrocytes (OLs) in a spontaneous co-culture which also includes astrocytes. This model physiologically resembles the thyroid hormone-mediated myelination and myelin repair process, allowing the addition of pathological interferents which model disease mechanisms. We show how to mimic the two main components of demyelinating diseases (i.e., hypoxia/ischemia and inflammation), recreating their effect on developmental myelination and adult myelin repair and taking all the cell components of the system into account throughout, while focusing on differentiating OPCs.

This spontaneous mixed model, coupled with cell-based high-content screening technologies, allows the development of a robust and reliable drug screening system for therapeutic strategies aimed at combating the pathological processes involved in demyelination and at inducing remyelination.

Introduction

In the central nervous system (CNS), myelin forming cells (oligodendrocytes, OLs) and their precursors (oligodendrocyte precursor cells, OPCs) are responsible for developmental myelination, a process which occurs during the peri- and post-natal periods, and for myelin turnover and repair (remyelination) in adulthood¹. These cells are highly specialized, interacting anatomically and functionally with all the other glial and neuronal components, making them a fundamental part of CNS structure and function.

Demyelinating events are involved in different CNS injuries and diseases², and mainly act on OPCs and OLs by way of multifactorial mechanisms, both during development and adulthood. The undifferentiated precursors are driven by differentiating factors, mainly thyroid hormone (TH), in a synchronized process³ which leads the OPC to recognize and respond to specific stimuli which induce proliferation, migration to the non-myelinated axon, and differentiation into mature OLs which in turn develop the myelin sheath⁴. All these processes are finely controlled and occur in a complex environment.

Due to the complex nature of myelination, remyelination and demyelination events, there is a great need for a simplified and reliable in vitro method to study the underlying mechanisms and to develop new therapeutic strategies, focusing on the main cellular player: the OPC⁵.

For an in vitro system to be reliable, a number of factors need to be taken into account: the complexity of the cellular environment, age-related cell-intrinsic differences, physiological TH-mediated differentiation, pathological mechanisms, and the robustness of the data⁶. Indeed, the unmet need in the field is a model which mimics the complexity of the in vivo condition, not successfully achieved through the use of isolated pure OPC cultures. In addition, the two main components of demyelinating events, inflammation and hypoxia/ischemia (HI), directly involve other cell components that may indirectly affect the physiological differentiation and maturation of OPCs, an aspect which cannot be studied in over-simplified in vitro models.

Starting from a highly predictive culture system, the subsequent and more general challenge is the production of robust and reliable data. In this context, cell-based high-content screening (HCS) is the most suitable technique⁷, since our aim is firstly to analyze the entire culture in an automatic workflow, avoiding the bias of choosing representative fields, and secondly to obtain the automatic and simultaneous generation of imaging-based high-content data⁸.

Given that the main need is to achieve the best balance between in vitro simplification and in vivo-mimicking complexity, here we present a highly reproducible method for obtaining OPCs derived from neural stem cells (NSCs) isolated from the fetal forebrain and the adult sub-ventricular zone (SVZ). This in vitro model encompasses the entire OPC differentiation process, from multipotent NSC to mature/myelinating OL, in a physiological TH-dependent manner. The resulting culture is a dynamically differentiating/maturating system which results in a spontaneous co-culture consisting mainly of differentiating OPCs and astrocytes, with a low percentage of neurons. This primary culture better mimics the complex in vivo environment, while its stem cell derivation allows simple manipulations to be performed to obtain the cell lineage enrichment desired.

On the contrary to other drug screening strategies using cell lines or pure cultures of primary OPCs, the method described here allows the study of the effect of pathological interferents or therapeutic molecules in a complex environment, without losing the focus on the desired cell type. The HCS workflow described permits an analysis of cell viability and lineage specification, as well as lineage-specific cell death and morphological parameters.

Protocol

All animal protocols described herein were carried out according to European Community Council Directives (86/609/EEC) and comply with the guidelines published in the NIH Guide for the Care and Use of Laboratory Animals.

1. Solutions and reagents

1. Prepare standard medium: DMEM/F12 GlutaMAX 1x; 8 mmol/L HEPES; 100 U/100 μg Penicillin/Streptomycin (1% P/S); 1x B27; 1x N-2.
2. Prepare neurosphere medium: add 10 ng/mL bFGF; 10 ng/mL EGF to standard medium.
3. Prepare oligosphere/OPC medium: add 10 ng/mL bFGF; 10 ng/mL PDGF-AA to standard medium.
4. Prepare oligodendroctye differentiation medium: add 50 nM T3; 10 ng/mL CNTF; 1x N-acetyl-L-cysteine (NAC) to standard medium.
5. Prepare non-enzymatic dissociation buffer: add 1% P/S to non-enzymatic dissociation buffer and keep ice cold.
6. Prepare sucrose solution: HBSS, 0.3 g/mL sucrose.
7. Prepare BSA washing solution: EBSS, 40 mg/mL BSA, 0.02 mL/l HEPES.
8. Prepare enzymatic dissociation buffer: HBSS, 5.4 mg/mL D-glucose, 15 mmol/L HEPES, 1.33 mg/mL Trypsin, 0.7 mg/mL Hyaluronidase, 80 U/mL DNase.
9. Prepare cytokine mix: TGF-β1, TNF-α, IL-1β, IL-6, IL-17, and IFN-γ (20 ng/mL each).
10. Prepare cytokine mix vehicle: 0.04% of the stock (10% glycerol/100 nM glycine/25 nM Tris, pH 7.3).
11. Prepare oxygen-glucose deprivation medium: standard medium using DMEM w/o glucose. Depending on the stringency of the desired glucose deprivation condition, it is possible to remove also B27 and/or N2 from the medium to avoid glucose-related compounds (e.g., D-Galactose in B27).

2. Dissection and NSC isolation

NOTE: Fetal and adult NSCs were isolated from E13.5 fetal forebrain or 2.5-month-old adult sub-ventricular zone (SVZ), following the Ahlenius and Kokaia protocol⁹ with modifications.

1. Fetal NSC cultures
   NOTE: Before starting the dissections, prepare 1.5 mL tubes containing 150 µL of non-enzymatic dissociation buffer each; clean Petri dishes and add ice cold HBSS.
   1. Collect the embryos at E13.5 - 14.5 from timed pregnant mice and place in a Petri dish containing cold HBSS.
   2. Decapitate the embryos using forceps.
   3. Place the heads of the embryos in a clean Petri dish containing ice cold PBS and remove the skin from the skull with forceps, using magnifying glasses or a stereoscope.
   4. Once the brain is visible and cleared of skin, squeeze it out by applying pressure at the sides with forceps.
   5. Remove the cerebellum, keep only the forebrain and remove the meninges with forceps.
   6. Place the isolated tissue in the non-enzymatic dissociation buffer and repeat the dissection steps with the other embryos. Insert the tissue from 2–3 animals into each tube containing the buffer.
   7. Incubate at 37 °C for 15 min under continuous shaking.
   8. After incubation, add 850 µL of standard medium and mix by pipetting until the suspension is free of clumps.
   9. If non-dissociated tissue is still visible, wait for 2 min at RT until it deposits at the bottom of the tube.
   10. When dissociation is complete, count the cells and plate them in suspension at a density of 10–50 cells/µL in a T-25 or T-45 flask containing 10–30 mL of neurosphere medium, kept in a vertical position to avoid cell adhesion.
2. Adult NSC cultures
   1. Sacrifice animals by cervical dislocation.
   2. Collect brains from 4–5 mice in a 50 mL tube containing ice cold HBSS.
   3. Place the brain on a cold sterile surface. For this purpose, use a T-25 flask filled with water and placed at -20 °C overnight. At the time of the experiment, cover the flask with sterile aluminum foil.
   4. Place the brain ventral side downwards, in rostro-caudal direction, and remove the olfactory bulbs using a razor blade.
   5. Using a razor blade, cut 2–3 coronal slices of 1 mm thickness, from the cortex to the optical chiasma.
   6. Place the slices on the cold surface in a ventro-dorsal position and identify the corpus callosum and the two lateral ventricles.
   7. Using magnifying glasses or a stereoscope, isolate the walls of the lateral ventricles, taking care not to carry pieces of the corpus callosum.
   8. Put the isolated tissue in the enzymatic dissociation buffer (5–10 mL) and incubate at 37 °C for 15 min.
   9. Mix the solution, pipetting several times (at least 50), and incubate again at 37 °C for 10 min.
   10. Neutralize the trypsin by adding 5 mL of standard culture medium and filter the solution using a 70 µm filter.
   11. Centrifuge the filtered solution for 5 min at 400 x g.
   12. Resuspend the pellet in the sucrose solution and centrifuge for 10 min at 500 x g.
   13. Resuspend the pellet in BSA washing solution and centrifuge for 7 min at 400 x g.
   14. Resuspend the pellet in the standard culture medium, count the cells, and perform plating as described above (in step 2.1.10).

3. Primary neurospheres

1. Add the growth factors (bFGF/EGF) every 2 days.
2. Every 4–6 days (depending on cell density), change half of the medium as follows:
   1. Transfer the entire cell suspension to a 15 or 50 mL tube.
   2. Centrifuge for 5 min at 400 x g.
   3. Remove half of the volume.
   4. Add the same amount of fresh medium, gently mix by pipetting, and add growth factors.

4. Oligospheres

NOTE: Oligodendrocyte differentiation is performed following the Chen protocol¹⁰ with modifications.

1. When the neurospheres reach a diameter of 100–150 µm, they are ready to be passed. To do so, transfer the entire cell suspension to a 15 or 50 mL tube, and centrifuge for 5 min at 400 x g.
   1. Rapidly evaluate the diameter by taking pictures of the spheres using an inverted transmitted light microscope and opening them by ImageJ software.
   2. Click on the Analyze menu and from the Tools window, select Scale bar.
   3. Set 150 µm as Width in microns and compare the scale bar with the spheres.
2. Remove the entire volume by inversion and resuspend the pellet in 180 µL of fresh standard culture medium. Pipette 50 times to allow disaggregation of the spheres.
3. Add 810 µL of fresh standard culture medium, count the cells, and re-plate them as described for the neurospheres.
4. Add bFGF/PDGF-AA 10 ng/mL every 2 days.
5. Every 4–6 days (depending on cell density), change half of the medium as follows:
6. Transfer the entire cell suspension to a 15 or 50 mL tube.
7. Centrifuge for 5 min at 400 x g.
8. Remove half of the volume.
9. Add the same amount of fresh medium, gently mix by pipetting, and add growth factors.

5. Plate coating

1. Poly-D,L-ornithine/laminin coating: at least 2 days before plating the OPCs, add 50 µg/mL poly-D,L-ornithine solution, diluted in PBS, to each well (40 µL/well for 96-well plates) and incubate at RT overnight.
2. The following day, remove the liquid and wash three times with distilled sterile water.
3. Let the plates dry at RT overnight. The following day, add a laminin solution diluted in PBS (5 µg/mL; 40 µL/well for 96-well plates) and incubate for 2 h at 37 °C.

6. Cell seeding

1. When the oligospheres reach a diameter of 100–150 µm, they are ready to be dissociated and seeded on the poly-D,L-ornithine/laminine coated plates. To do so, transfer the entire cell suspension to a 15 or 50 mL tube, and centrifuge for 5 min at 400 x g (as indicated in step 4.1)
2. Remove the entire volume by inversion and resuspend the pellet in 180 µL of fresh standard culture medium. Pipette 50 times to allow disaggregation of the spheres.
3. Add 810 µL of fresh standard culture medium and count the cells.
4. Remove the laminin solution from the wells and plate the cells at 3,000 cell/cm² density (100 µL/well for 96-well plates).

7. OPC differentiation induction

1. After 3 days, remove the entire medium and add the same volume of oligodendrocyte differentiation medium.
2. Change half of the medium every 4 days and add fresh differentiation mix (T3/CNTF/NAC) every 2 days.

8. Induction of inflammation-mediated differentiation block

1. After neurosphere dissociation and oligosphere production (section 4), add the cytokine mix to the culture medium and keep oligospheres exposed to cytokines for the whole spheres formation step.
   NOTE: The volume depends on the number of cells, since for the spheres forming cells are seeded at 10–50 cells/µL.
2. If the medium needs to be changed, change the entire volume and add the cytokine mix once more.

9. Induction of oxygen-glucose deprivation cell death

1. At -1 DIV (2 days after cell seeding in multiwell plates), remove the medium and conserve it in a new multiwell plate.
2. Add half the volume (50 µL for 96-well plates) of OGD-medium (OGD group) or fresh medium (control group). The half amount of volume is used to reduce the exchange of oxygen between the liquid and the air.
3. Place the OGD group cultures in an airtight hypoxia chamber saturated with 95% N₂ and 5% CO₂. To achieve saturation of the chamber, let the gas mixture flow for 6 min at 25 l/min before closing the chamber pipes.
4. Incubate the hypoxic chamber in the incubator for 3 h. The control group and plates containing the medium removed and conserved at step 9.1 should also be left in the incubator.
5. Remove the glucose-free (OGD group) or the new medium (control group) and add the medium removed and conserved at step 9.1.

10. Immunocytochemistry

1. At the desired time point, fix the cells with cold 4% paraformaldehyde for 20 min at RT.
2. Wash twice with PBS (10 min of incubation for each wash at RT).
3. Incubate 1 h at RT with the blocking solution (PBS Triton 0.3% containing 1% BSA and 1% donkey/goat normal serum).
4. Incubate with primary antibody mix (Table 1), diluted in PBS triton 0.3%, overnight at 4 °C.
5. Wash twice with PBS (10 min of incubation for each wash at RT).
6. Incubate with secondary antibody (Table 1) solution diluted in PBS triton 0.3% adding Hoechst 33258 for 30 min at 37 °C.
7. Wash twice with PBS (10 min of incubation for each wash at RT).

11. HCS analysis of cell viability, lineage composition, and lineage-specific cell death

NOTE: The HCS representative images and workflow are shown in Figure 2A,B.

1. Select the Compartmental Analysis algorithm from the main menu of the software (HCS Studio v 6.6.0) and select Scan from the main menu Develop Assay/Scan Plate.
2. In the iDev window, select New and then select General intensity Measurement Tool from the Develop Assay template.
3. Click on Create on the right side of the menu, selecting the 10x objective.
4. This will open the Configure Acquisition menu. In this window, select the following parameters: (a) number of channels: the first one for the Hoechst nuclear staining (BGRFR_386) and one for each lineage-specific marker used in the reaction (b) select software focus on channel 1 and autofocus interval as 1 (c) select the plate model from the list.
5. From the acquisition menu, look at the quality of the staining in different wells and different fields and manually select the exposure time by selecting Fixed Exposure Time in the menu.
6. Once the acquisitions parameters are set, select Mini Scan on the top of the menu and select ten fields per well in two wells per experimental condition. This will allow the set-up of all the analysis parameter in a subset of fields for the entire plate.
7. When the mini scan is finished, click on the Configure Assay Parameter to configure the algorithm of the analysis.
8. Click on Configure Groups in the right side of the window and drag-and-drop the wells of the miniscan. Click on the Add button in the Groups sub-section to configure the different groups.
9. Follow the workflow on the left side of the window step by step to develop the whole algorithm. First select Process Image for each channel and click on Background Removal and on the desired level.
10. First identify and select the nuclei by nuclear staining. Click on Identify Primary Object – Channel 1 to select the real nuclei and avoid analyzing artifact and debris. For this purpose, zoom in a representative picture of nuclear staining and check whether the nuclei are well surrounded by the perimeter built by the software. It is possible to change the thresholding value and to apply segmentation algorithms to better identify single nuclei.
11. Once the nuclei are defined correctly, click on the following step: Validate Primary Object. Select Object.BorderObject.Ch1 to avoid the analysis of nuclei at the border of each field image. Select Object.Area.Ch1 and, by moving the “low” and “high” bars on the histograms, remove all the identified debris or big-objects corresponding to aggregates or artifacts.
12. Check all the mini scan representative images of all the experimental conditions to be sure that the selected parameters fit with all of them.
13. Click on Identify Spots for each channel corresponding to the specific lineage markers, and select the Ring values: Width = 3 and Distance = 0. This will allow the identification of the cytoplasmatic fluorescence. According to the cell density, these values can be adapted. The software will automatically avoid the overlapping between adjacent rings.
14. Select Reference Levels in the workflow to build the analysis. The setting of the reference levels will allow the automatic counting of condensed nuclei, based on the nuclear size and nuclear staining intensity, and of specific marker-positive cells, based on the cytoplasmatic fluorescence identified by the Ring.
15. First click on Object.Area.Ch1. In the mini scan images, select a condensed nucleus and move the “LOW” bar on the histograms in order to select as “condensed” all the nuclei under this size.
16. Click on Object.AvgIntensity.Ch1. In the mini scan images, select a condensed nucleus and move the “HIGH” bar on the histograms in order to select as “condensed” all the nuclei above this fluorescence intensity.
17. Click on Object.RingAvgIntensity for each channel of lineage specific markers. Select in your mini scan images a positive cell and move the “HIGH” bar on the histograms in order to select as “positive” all the cells above this fluorescence intensity.
18. Check all the mini scan representative images of all the experimental conditions to be sure that the selected parameters fit with all of them.
19. On the top menu, select Population Characterization and select Event Subpopulation.
20. As Type 1 Event, select ObjectAreaCh1 on the left list, then click on the AND > button and finally select ObjectAvgIntensityCh1. This will allow the identification of condensed nuclei, as a combination of low area and high intensity.
21. In the same window, deselect all the Scan Limits.
22. Click on Select Features to Store in the top menu, to choose the parameters to keep in the analysis.
23. Select Well Features and move from the left list to the right only the desired parameters: (a) SelectedObjectCountPErValidField (b) %EventType1ObjectCount (c) %High_RingAvgIntensity (For each channel of the specific lineage markers).
    NOTE: This analysis will give as readout the total number of cells, the percentage of condensed nuclei, and the percentage of lineage-specific positive cells for each analyzed marker on the total cell number. If the percentage of the different lineages is needed only on live cells, is it possible either to keep the value “High_RingAvgIntensity” for the channel (absolute number of positive cells) and recalculate the percentage on total cell numbers after the subtraction of the percentage of dead cells.
    1. Alternatively, it is possible to remove the dead cells from the analysis setting the same parameters used to identify condensed nuclei (steps 11.14–11.15) on the nuclei validation (step 11.11).
24. Select Scan Plate from the main top menu and click on the plate symbol on Scan Setting sub-menu on the top section to identify the well to analyze.
25. Write the name of the experiment and the description and once all the settings are completed, press the play symbol.

Results

The first phase of the culture may vary in duration, depending on seeding density and on whether the spheres are of fetal or adult origin. Moreover, oligospheres display a reduced population doubling compared to neurospheres (Figure 1B). Moreover, spheres production from adult tissue is slower and it may take 2–3 weeks to generate oligospheres compared to fetal that may take 1–2 weeks, depending on the seeding density.

Once seeded, the entire differentiation phase...

Discussion

The complex nature of myelination/remyelination processes and demyelinating events makes the development of predictive in vitro systems extremely challenging. The most widely used in vitro drug screening systems are mostly human cell lines or primary pure OL cultures, with increasing use of more complex co-cultures or organotypic systems¹⁵. Even if such systems are coupled with high content technologies, pure OL cultures remain the method of choice when developing screening platforms

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
96-well plates - untreated	NUNC	267313
B27 supplement (100x)	GIBCO	17504-044
basic Fibroblast Growth Factor (bFGF)	GIBCO	PHG0024
BSA	Sigma-Aldrich	A2153
Ciliary Neurotropic Factor (CNTF)	GIBCO	PHC7015
DMEM w/o glucose	GIBCO	A14430-01
DMEM/F12 GlutaMAX	GIBCO	31331-028
DNase	Sigma-Aldrich	D5025-150KU
EBSS	GIBCO	14155-048
Epidermal Growth Factor (EGF)	GIBCO	PHG6045
HBSS	GIBCO	14170-088
HEPES	GIBCO	15630-056
Hyaluronidase	Sigma-Aldrich	H3884
IFN-γ	Origene	TP721239
IL-17A	Origene	TP723199
IL-1β	Origene	TP723210
IL-6	Origene	TP723240
laminin	GIBCO	23017-051
N-acetyl-L-cysteine	Sigma-Aldrich	A9165
N2 supplement (50x)	GIBCO	17502-048
Non-enzymatic dissociation buffer	GIBCO	13150-016
PBS	GIBCO	70011-036
Penicillin / Streptomycin	Sigma-Aldrich	P4333
Platelet Derived Growth Factor (PDGF-AA)	GIBCO	PHG0035
poly-D,L-ornitine	Sigma-Aldrich	P4957
TGF-β1	Origene	TP720760
TNF-α	Origene	TP723451
Triiodothyronine	Sigma-Aldrich	T2752-1G
Trypsin	Sigma-Aldrich	T1426