1
Title:
Isolation, purification and expansion of myelination-competent neonatal mouse
Schwann Cells
Running Title:
Myelination-competent mouse Schwann Cells
Authors:
Henrika Honkanen1, Outi Lahti1, Marja Nissinen2, Riina M. Myllylä2, Satu
Päiväläinen1, Maria H. Alanne2, Sirkku Peltonen3, Juha Peltonen2 & Anthony M.
Heape1
Affiliations:
1
Myelin Group, Department of Anatomy & Cell Biology, University of Oulu, Aapistie
7A, 90014 Oulu, Finland.
2
NF1 Research Team, Department of Anatomy & Cell Biology, University of Oulu,
Aapistie 7A, 90014 Oulu, Finland.
3
Cell-Cell Interaction Group, Department of Dermatology, University of Turku,
Savitehtaankatu 1, 20521 Turku, Finland.
Corresponding author:
Doc. Anthony M. Heape,
The Myelin Group,
Department of Anatomy & Cell Biology,
P.O. Box 5000 (Aapistie 7A),
90014 University of Oulu,
Finland.
E-mail: Anthony.Heape@oulu.fi
Phone: (+358)-8-5375197
Fax:
(+358)-8-5375172
Funding
This work was funded by the NeoBio Research Program of the Finnish National
Technology Agency (Tekes), and a research grant from the Academy of Finland. AMH is a
Research Fellow of the Academy of Finland.
2
Abstract
Most studies of PNS myelination using culture models are currently performed using
dorsal root ganglion neurons (DRGN) and Schwann cells (SC) pre-purified from the rat.
However, the potential of the model is severely compromised by the lack of rat myelin
mutants, and the published protocols work very poorly with cells from mice, for which
numerous myelin mutants are available. Here, we describe the isolation, purification and
expansion of wild-type, myelination-competent SCs from the sciatic nerves of 4-day-old
mouse pups.
The protocol consistently provides ~95 %-pure mouse SC yields of 1.9 - 3.3 x 106 cells
from the sciatic nerves of 12 to 15 four-day-old mouse pups, within 14 - 20 days of the
isolation of the nerves. The proliferation rate of the cultured mouse SCs, expressed as
“fold growth/week”, ranges from 2.7 to 4.30 under optimal conditions. Mouse SC
proliferation usually ceases within 4 weeks, when the cells become quiescent. Growth is
re-induced by the presence of sensory neurons; neuregulin is not sufficient for this effect.
The SCs isolated by this protocol maintain their ability to form compact myelin in culture,
as judged by the segregated expression patterns of early (myelin-associated glycoprotein)
and late (myelin basic protein) markers of myelination in a three-dimensional DRGN/SC
coculture model. The SC yields are sufficient to perform 100 to 150 individual myelinating
DRGN/SC coculture assays.
Key words
Rodent; Peripheral nervous system; Glial cell; Growth; Myelin; Culture; Method.
3
Introduction
Schwann cells (SC) are the myelin-forming glial cells of the peripheral nervous system
(PNS). As they are the direct and indirect targets of numerous hereditary and acquired
peripheral myelin diseases affecting Man, there is strong interest in the research
community to understand their biology and pathology. This strong interest has, over the
course of the last few decades, led to the establishment of numerous animal models of
human SC-related diseases and culture systems that have proved invaluable in studies of
cellular and molecular processes related to SC biology and, in particular, to myelin
formation, maintenance and disease.
Most studies using myelinating culture models are currently performed using dorsal root
ganglion neurons (DRGN) and SCs pre-purified from the rat (Brockes et al. 1979; Kleitman
et al. 1997). This model has been employed successfully by many research groups to
study various aspects of PNS myelination. However, the potential of the model is severely
compromised by the lack of rat myelin mutants. A notable exception is the rat model of
human Charcot-Marie-Tooth type 1A disease (Serada et al. 1996; Nobbio et al. 2006).
In contrast, there are several spontaneous mouse myelin mutants, and an ever-growing
number of bio-engineered mutants. If employed in a DRGN/SC coculture model, these
mutants would be highly valuable for elucidating cell type-specific biomolecular and
cellular events taking place during normal and pathological myelination. In particular, when
used with mouse DRGNs and SCs this model would provide the unique opportunity of
performing mixed phenotype/genotype cocultures, exploiting the abundance of mice with
spontaneous or bio-engineered mutations expressed by their SCs, or sensory neurons.
Unfortunately, the published protocols employed for the rat myelinating coculture model
with prepurified SCs and neurons work very poorly with mouse cells, a situation that is
largely, but not exclusively, related to the difficulties of isolating and purifying sufficient
quantities of myelination-competent mouse SCs. Consequently, for culture-based studies
of mouse PNS myelination, researchers tend to resort to the somewhat less refined, and
cellularly heterogeneous (but see Kim et al. 1997) dorsal root ganglion (DRG) explant
culture model, in which the neurons and the SCs are necessarily derived from the same
animal (for example; Liu et al. 2005).
Here, we describe a protocol for the isolation, purification and expansion of myelinationcompetent SCs from 4-day-old mouse pups, in quantities sufficient for carrying out
medium-to-large series of myelinating cocultures (i.e. in excess of 100 cocultures).
4
Materials and Methods
Culture media
The compositions of the culture media employed in this study are as follows:
Basic growth medium: High-glucose Dulbecco’s modified essential medium (HG-DMEM;
Sigma), containing 10 % inactivated horse serum (Gibco), 4 mM L-glutamine (Sigma), 100
units (u)/ml penicillin/streptomycin (Sigma), 2 ng/ml human heregulin-β1 (Sigma), and 0.5
μM forskolin (Sigma).
SC Growth Medium: HG-DMEM, containing 10 % inactivated horse serum, 4 mM Lglutamine, 100 u/ml penicillin/streptomycin, 2 ng/ml human heregulin-β1, 0.5 μM forskolin,
10 ng/ml human basic fibroblast growth factor (bFGF; Sigma), and 20 μg/ml bovine
pituitary extract (PE; Sigma).
HMEM: HG-DMEM, containing 20 mM Hepes, 10 % inactivated horse serum, 4 mM Lglutamine, and 100 u/ml penicillin/streptomycin.
Complement-mediated Cytolysis Medium: HMEM containing 4 µg/ml anti-mouse CD90
anti-Thy-1.2 (Serotec).
DRG Growth Medium: Minimum essential medium with Earl’s salts (EMEM; Sigma),
containing 10 % inactivated horse serum, 4 g/L glucose and 50 ng/mL nerve growth factor
(NGF).
Differentiation Medium: HG-DMEM/Hams F12 (Gibco) (1:1, vol/vol), containing 1 % N2supplement (Gibco), and 50 ng/ml nerve growth factor (NGF; R&D Systems).
Myelination Medium: EMEM, containing 5 % inactivated horse serum (Gibco), 0.4 % D-(+)glucose (Sigma), 4 mM L-glutamine, 50 ng/ml NGF, 0.5 μM forskolin, 20 μg/mL PE, 0.25%
N2-supplement, and 50 µg/ml ascorbic acid (Sigma).
Preparation of poly-L-lysine-, collagen-, and laminin-coated culture surfaces
All coating solutions (0.25 mg/ml collagen type 1 in 0.1 N acetic acid; 10 µg/ml laminin
from Engelbreth-Holm-Swarm murine sarcoma basement membrane; and 10 µg/ml poly-Llysine) were prepared according to the instructions of the manufacturer (Sigma).
Collagen type 1 was applied to sterile 35-mm diameter (Ø) tissue culture dishes (Greiner
Bio-One) for 3 h at +37 °C, and the surface was washed 3 times with sterile H2O.
Laminin was applied to 35-mm (Ø) tissue culture dishes for 2 h at +37 °C, and the surface
was washed 3 times with sterile H2O.
5
Poly-L-lysine was applied to 35-, or 60-mm (Ø) tissue culture dishes, or 13-mm (Ø) glass
coverslips, for 45 min at room temperature, and the surface was washed twice with sterile
H2O.
Animals and isolation of nerve tissue
The use of the mice and the protocols employed for the isolation of the tissues were
approved by the Animal Care and Use Committee of the University of Oulu, Finland
(permit No. 035/04).
The protocol employed for the isolation and growth of mouse SCs is based on the method
of Brockes et al. (1979), as modified by Kleitman et al. (1997), for the rat SCs.
Four-day-old CD-1 mouse pups (12 - 15 pups/preparation, unless stated otherwise), raised
in the Laboratory Animal Centre (University of Oulu, Finland), are sacrificed by CO2
exposure followed by decapitation, and cleaned by brief submersion in 70 % ethanol. Both
sciatic nerves are exposed, dissected out under a binocular microscope, and placed in a
sterile culture dish (3-cm Ø, Greiner Bio-One) containing ice-cold phosphate-buffered
saline (PBS) without CaCl2 or MgCl2 (Sigma), where they are maintained until the nerves
have been harvested from all of the mice. The nerves are then carefully stripped of
connective tissue with fine forceps and transferred to a new culture dish containing 3.0 ml
ice-cold PBS. The cleaned nerves are shredded with forceps until they resemble cotton
balls. The sciatic nerves are kept on ice for not more than 1 hour.
Dissociation and plating of nerve fragments
The shredded nerves, together with the PBS (3 ml), are transferred from the culture dish to
a 15-mL plastic Falcon tube containing 0.5 ml 2.5 % trypsin (Sigma), 0.5 ml 10 mg/ml
collagenase A (Roche), and 3 ml PBS. The culture dish is rinsed with 3 ml of PBS, and the
rinsate is added to the Falcon tube containing the nerve/enzyme mixture (total volume 10
mL; final enzyme concentrations: 0.125 % trypsin and 0.05 % collagenase A). The
enzymatic digestion is allowed to proceed for 30 min at +37 °C, mixing occasionally, after
which the mixture is centrifuged at 190 x g for 5 min in a benchtop centrifuge (Centra CL2
Thermo IEC). The supernatant is removed, taking care not to aspirate the fluffy nerve
fragments, and the residual enzymes are flushed from the cellular pellet by 3 consecutive
washes with 7 ml of high-glucose DMEM (Sigma) containing 10 % horse serum (Gibco)
and 5-min centrifugations at 190 x g. The cells are resuspended in 2 ml of Basic Growth
Medium (see above), plated on a poly-L-lysine-coated 60-mm plastic tissue culture dish,
and placed in a cell culture incubator (Forma Scientific) at + 37 °C, with a 5 % CO2
6
humidified atmosphere. Two days later, the Basic Growth Medium, including any material
that has not already adhered to the culture dish, is transferred to a fresh poly-L-lysinecoated 35-mm tissue culture dish. Two mL of fresh Basic Growth Medium are added to the
original culture dish containing the attached cells, and both dishes are replaced in the cell
culture incubator for a further 2 days.
Purification of Schwann cells – complement-mediated cytolysis
Complement-mediated cytolysis of the contaminating fibroblasts with mouse anti-mouse
CD90 anti-Thy-1.2 antibody (Dong et al. 1999) is performed 96 h (4 days) after the
enzymatic dissociation and plating of the nerve fragments. The Basic Growth Medium is
removed from the cultures and the attached cells are rinsed once with Ca2+- and Mg2+-free
Hank’s balanced salt solution (HBSS) (Gibco) in 20 mM Hepes Buffer (Cambrex), and
once with HMEM (see above), before adding 1 ml of the Complement-mediated Cytolysis
Medium (see above). After incubation for 15 min at +37 °C, 200 µl rabbit HLA-ABC
complement sera (Sigma) is added and the incubation is continued for another 2 h at +37
°C. Cytolysis is terminated by rinsing the cells twice with Ca2+- and Mg2+-free HBSS in 20
mM Hepes Buffer.
The complement-mediated cytolysis is repeated before the first passage (see below) if
fibroblasts are still visible in the cultures. Fibroblasts are routinely identified by phasecontrast microscopy as flattened polymorphic cells with large, round, non-luminescent
nuclei.
Expansion of Schwann cells
After the complement-mediated cytolysis of the fibroblasts, 2 mL SC Growth Medium (see
above) are added to each culture and the plated cells are maintained in a cell culture
incubator (Forma Scientific) at +37 °C, with a 5 % CO2 humidified atmosphere, changing
the medium every 2 days. The cells are passaged when the dishes are ~80 % confluent,
i.e. after 7-12 days in culture, depending on the initial SC population size and the cell
proliferation rate.
For passaging, the cells are rinsed two times with Ca2+- and Mg2+-free HBSS in 20 mM
Hepes buffer, followed by a brief incubation (less than 2 minutes) with ~0.5 mL 1 x trypsin
versene (Sigma). Trypsination is stopped by the addition of 1-2 mL of horse serum,
followed by rinsing with high-glucose DMEM containing 10 % horse serum. After
centrifugation (190 x g for 5 min), the cells are again rinsed with 5-10 mL high-glucose
7
DMEM containing 10 % horse serum, recentrifuged and suspended in 1.0 mL of SC
Growth Medium, and counted in a Bürker-Türk cell counting chamber.
The SCs, resuspended in SC Growth Medium, are replated on poly-L-lysine-coated culture
dishes at a density of 5 - 6 x 104 cells/ 35-mm dish (with 1 mL medium), or 1.5 – 1.9 x 105
cells/60-mm dish (with 2 mL medium), and grown at +37 °C, with a 5 % CO2 humidified
atmosphere, changing the medium every 2 days, until the cultures reach ~80 %
confluence. The cells are then trypsinated and counted, as described above.
For frozen storage, the trypsinated cells are centrifuged, resuspended in foetal bovine
serum (FBS; Gibco) containing 7% dimethylsulfoxide (DMSO; Sigma), and stored in liquid
nitrogen.
Characterization of the Schwann cell cultures
Phase-contrast microscopy. Routinely, SCs are identified as elongated bi-, or tripolar cells,
with an oval luminescent soma. Fibroblasts appear as flattened, polymorphic, nonluminescent cells with a large, round nucleus.
Immunofluorescence microscopy. SCs were identified by triple fluorescence microscopy
as follows. Cells harvested after one complement-mediated cytolysis and two subculture/trypsination cycles, performed as described above, were plated in SC Growth
Medium on poly-L-lysine-coated glass coverslips (Ø 13.0 mm) in 35-mm culture dishes, at
a density of 20000 cells/coverslip, and cultured for 48 h, as described above. The culture
medium was then removed and, after rinsing the coverslips with PBS, the cells were fixed
with 2.5 % para-formaldehyde for 10 min at room temperature, permeabilized with ice-cold
methanol for 6 min, and again rinsed with PBS. Non-specific binding surfaces were
blocked by incubation for 15 min, at room temperature, with 1 % bovine serum albumin
(BSA) in PBS.
Primary antibodies (Abcam): mouse monoclonal anti-S100β antibody and rabbit polyclonal
anti-Glial Fibrillary Acidic Protein (GFAP), were mixed and diluted in PBS containing 1 %
BSA, to final dilutions of 1:200 and 1:1000, respectively.
Secondary antibodies (Molecular Probes): Alexa Fluor® 488-conjugated goat anti-mouse
IgG (H+L) and Alexa Fluor® 568-conjugated goat anti-rabbit IgG (H+L), were mixed and
diluted in PBS containing 1 % BSA, both to final dilutions of 1:100.
The fixed cells, on the coverslips, were incubated with the primary antibody mixture for 45
min at +4 °C, washed 3 x 5 min with PBS containing 1 % BSA, incubated with the
secondary antibody mixture for 30 min at + 4 °C, and washed once with PBS. Finally, the
8
cells were incubated with 0.2 µg/ml Hoechst nucleus dye (Sigma) in PBS for 10 min,
washed once with PBS and 3 times with H2O, and then mounted on a glass microscope
slide. The slides were observed under a Nikon Eclipse E600 fluorescence microscope,
photographed using a QImaging MicroPublisher 5.0 RTV camera, and analyzed with
QCapture PRO Software.
The total number of cells (Hoechst-positive), the number of SCs (S100β-positive, GFAPpositive and Hoechst-positive), and the number of fibroblasts (Hoechst-positive, but
S100β- and GFAP-negative), were counted in 10 random fields (30-50 cells/field). The
abundance of each cell type is expressed relative to the total number of Hoechst-positive
cells in each field.
Comparison of poly-L-lysine-, collagen- and laminin-coated growth surfaces
Following the purification of the SCs by complement-mediated cytolysis, the SCs were
replated, in SC Growth Medium, on 35-mm poly-L-lysine-coated tissue culture dishes and
grown to ~80 % confluency (7 days), as described above. The cells were then trypsinated,
counted and replated, in SC Growth Medium, on 35-mm poly-L-lysine-, collagen-, or
laminin-coated dishes, prepared as described above, at a density of 6 x 104 SCs/dish.
After 8 days, the cultures were photographed with a Nikon CoolPix 950 digital camera
mounted on a Leica DMIL phase-contrast microscope, and then trypsinated and counted,
as described above.
Triplicate assays were performed for each culture surface coating, and the whole
experiment was performed two times with SC batches prepared separately from two
different sets of mouse pups. Mean cell numbers and standard deviations were
determined for the triplicate assays of each SC batch grown on each substratum, and
Student’s two-tailed t-test was used to determine the significance of the differences
between the yields obtained on each of the substrata.
Dorsal Root Ganglion Neuron and Schwann cell Coculture
Mouse SCs were prepared as described above, with two post-cytolysis passages. The
isolation, purification and growth of the mouse DRGNs, and the three-dimensional mouse
myelinating DRGN/SC cocultures, were performed as described by Päiväläinen et al.
(submitted for publication).
Briefly, DRGNs isolated from E14.5 CD-1 mouse embryos, were cultured in sterile 96-well
plastic culture dishes (Costar) containing a thin layer (~20 μL/well) of Matrigel© (BD
Biosciences) prepared with a 1:3 dilution according to the manufacturer’s instructions. 1.5
9
x 104 SCs, resuspended in DRG Growth Medium (see above), were seeded onto each
DRGN culture. One day later, the medium was changed to Differentiation Medium (see
above) for one week, renewing the medium every 2 days. The medium was then changed
to Myelination Medium (see above), and the cocultures were continued for a further 3
weeks, renewing the medium every 2 days.
Routine controls included purified DRGNs and SCs cultured alone, under conditions
identical to those employed for the DRGN/SC cocultures, and concurrently with the latter.
For immunocytochemistry, the myelinating DRGN/SC cocultures were fixed, permeabilized
and blocked as described above for the SC cultures, and analyzed by fluorescence
confocal microscopy.
Primary antibodies: mouse monoclonal anti-myelin-associated glycoprotein (MAG;
Chemicon) and rabbit polyclonal anti-myelin basic protein (MBP: Garbay et al. 1988) were
mixed and diluted in PBS containing 1 % BSA, both to final dilutions of 1:150.
Secondary antibodies (Molecular Probes): Alexa Fluor® 488-conjugated goat anti-mouse
IgG (H+L) and Alexa Fluor® 568-conjugated goat anti-rabbit IgG (H+L), were mixed and
diluted in PBS containing 1 % BSA, to final dilutions of 1:150 and 1:200, respectively.
Incubations with primary and secondary antibodies and all washes were performed as
described above for the SCs. The immunostained cocultures in the 96-well plates (nonmounted, but overlaid with PBS) were observed and photographed using a Zeiss LSM 510
Laser Scanning microscope equipped with Argon (excitation 458 nm, 488 nm and 514 nm)
and HeNe (excitation 543 nm) lasers, using a 40x objective.
10
Results
Comparison of Mouse Schwann cell growth on different substrates
Triplicate cultures of SCs from two separate isolations were grown on each of the three
substrata (collagen type 1, poly-L-lysine and laminin), as described in Materials and
Methods. After 8 days, representative fields in each culture were photographed under
phase-contrast microscopy, and the cells were harvested and counted.
All three substrata supported mouse SC growth (Figure 1), but cells cultured on collagen
type 1 and poly-L-lysine exhibited significantly higher (p < 0.05) proliferation rates (~1.9
population doublings [PD]/week) than those grown on laminin (~1.1 PD/week), yielding
average population growths (both SC batches combined) of about 5.0-, 4.6- and 2.3-fold
during the 8 days in culture. Although the average yield was slightly higher on collagen
than on poly-L-lysine, the difference was not significant, and we selected poly-L-lysine for
all subsequent experiments. Both bipolar and tripolar SCs were observed on all three
substrata (Figure 2).
Immunocytochemical characterization of the mouse Schwann cell cultures
Purified SC cultures were characterized after one round of complement-mediated cytolysis
by triple fluorescence microscopy, using Hoechst nucleus stain (for total cells), and
antibodies to GFAP, an intermediate filament protein found in embryonic and nonmyelinating SCs (Jessen et al. 1984, 1990), and S100, a commonly used SC marker
(Hyden and McEwen 1966). Neither GFAP, nor S100, are expressed by fibroblasts.
Figure 3 shows a typical field observed in these cultures.
94 ± 0.08 % of the cells were GFAP-positive and 93 ± 0.01 % were S100β-positive (all of
the latter were also GFAP-positive).
Schwann cell proliferation
SCs are usually seen to emerge from the sciatic nerve fragments within two days of plating
(Figure 4A). At the first passage, following the first complement-mediated cytolysis (i.e. 712 days after plating the nerve fragments, when the cultures have reached ~80%
confluency, Figure 4B), the total yield (N1) was 796000 ± 104000 cells (mean [m] ±
standard deviation [SD], of 5 separate isolations (A - E); range: 700000 – 990000 cells;
see Table 1b).
The cells were replated at a density of 5 - 6 x 104 cells/dish (35-mm, poly-L-lysine-coated,
1 mL SC Growth Medium), or 15 – 19 x 104 cells/dish (60-mm, poly-L-lysine-coated, 2 mL
11
SC Growth Medium), and cultured in the presence of heregulin, forskolin, bFGF and PE for
a further 4 - 8 days (Figure 4C), as specified in Table 1a, before reharvesting and
counting. Culture variables were thus: culture dish surface area (962 mm 2, or 2827 mm2),
volume of medium (1, or 2 mL), culture time (4, 7, or 8 days) and cell plating density (48 67 cells/mm2). Note that the SCs from isolation C were assayed in both 35-mm (1 mL
medium) and 60-mm (2 mL medium) culture dishes. The results are reported in Table 1b.
All SC proliferation assays are done at least in triplicate (n = 3 - 5).
Employing the protocol described in this paper, we consistently obtain ~95 %-pure mouse
SC yields of 1.6 - 3.3 x 106 cells from the sciatic nerves of 12 - 15 four-day-old mouse
pups, within 14 - 20 days of the isolation of the nerves (Table 1b). The yields are not
significantly different when isolations are performed from 3-, or 5-day-old mouse pups (not
shown).
Like the total yield, the proliferation rate of the cultured mouse SCs, expressed as “fold
growth/week” (GW), varies from one isolation to another, ranging from 2.17 ± 0.59
(isolation B), to 4.30 ± 0.62 (isolation C). Overall, the GW values observed in the cultures
grown on 35-mm dishes (3.74 ± 0.85, n = 10) tended to be higher than those grown on 60mm plates (2.72 ± 0.77, n = 11).
Neuronal influence on Schwann cell proliferation
Mouse SC proliferation tends to slow considerably after 3 weeks in pure cultures, and
usually ceases within 4 weeks, when the cells become quiescent (results not shown, but
see Fig. 5).
The comparison of SC behaviors in DRGN/SC cocultures and in mouse SC cultures
without neurons, in a 3-dimensional Matrigel matrix, indicates that the mouse SCs can be
re-induced to proliferate by a factor of neuronal origin (Fig. 5). The SCs attach to the 3dimensional Matrigel substratum within one day of seeding in the DRG Growth Medium,
both in the absence (Figure 5C) and in the presence (Figure 5E) of DRGNs, after which
the medium is changed to Differentiation Medium. One week later, although the SC
morphology has changed, the number of cells does not appear to have increased
significantly in the absence of DRGNs (Figure 5D). In contrast, when the SCs are cultured
with DRGNs under otherwise identical conditions, there is a clear increase in SC number
(Figure 5F). The increase is not the result of SCs, or SC precursors, growing out of the
DRGs, as demonstrated when the latter are cultured alone (Figures 5A and 5B), under
conditions identical to those employed in the SC cultures and DRGN/SC cocultures shown
in Figures 5C-F.
12
Mouse myelinating DRGN/SC cocultures
We tested the ability of a batch of freshly isolated mouse SCs produced using our protocol
to synthesize mature myelin in cocultures with embryonic mouse DRGs.
No mature myelin was observed during the first 2 weeks after the initiation of myelination
by the addition of the Myelination Medium. However, by 3 weeks, numerous myelin
sheaths are visible by both immunofluorescence microscopy, which revealed the expected
segregation of MBP and MAG (Figures 6A-C) in compact myelin sheaths, and phasecontrast microscopy (Figure 6D). Similar results are obtained when the cocultures are
performed with frozen SCs (not shown).
13
Discussion
Several protocols have been developed for the isolation and purification of SCs from
embryonic (Kim et al. 1995, 1997; Dong et al. 1999), neonatal (Seilheimer and Schachner
1987; Dong et al. 1999; Nicholson et al. 2001; Lepore et al. 2003) and adult mice (Verdu et
al. 2000; Lepore et al. 2003; Manent et al. 2003; Pannunzio et al. 2005). The reported
purities of the SC populations obtained using these techniques range from 85% (Verdu et
al. 2000; Pannunzio et al. 2005), to 99% or more (Seilheimer and Schachner 1987; Kim et
al. 1997; Manent et al. 2003).
Most of the above-mentioned publications describing mouse SC isolation focused
principally on the SC purity and, with the exception of that described by Verdu et al. (2000)
for the expansion of adult mouse SCs (~100000 cells/cm nerve), data concerning the total
SC yields is conspicuously absent. None of the reports indicate whether the resulting
mouse SCs are fully myelination-competent in culture. In one study, however, axonal
ensheathment, basal lamina synthesis, and early myelination stage-specific protein
expression (myelin-associated glycoprotein, or MAG), but no “mature” myelin, were
observed in mouse DRGN/SC cocultures (Seilheimer et al. 1989) using SCs prepared by
the method of Seilheimer and Schachner (1987).
In this paper, we describe a protocol allowing regular total yields of more than 2 x 106
mouse SCs (~95 % pure), within two to three weeks, from the sciatic nerves of 12 - 15
four-day-old mouse pups. These yields are sufficient to perform 100-150 individual
myelination assays using a 96-well, 3-dimensional, DRGN/SC coculture model
(Päiväläinen et al. submitted for publication). We also demonstrate that the purified SCs
are fully myelination-competent when cocultured with embryonic mouse (E14.5) DRGNs,
expressing both early and late markers of myelination, and forming mature myelin. The
principal differences between this protocol and the original protocol developed for rat SCs
are summarized in Table 2.
Mouse Schwann cell isolation, purification and proliferation
Mouse SCs not only appear to be highly sensitive to the relatively harsh isolation and
purification protocols conventionally employed for rat SCs (Kleitman et al. 1997), but they
also appear to be subject to a 3-week growth limit, after which they usually enter into a
quiescent state that is only lifted when the SCs are put in the presence of neurons. The
sensitivity to the established isolation and purification protocols could be satisfactorily
addressed by making small changes to the latter (see below). On the other hand, although
the results obtained in the course of this study provide clues as to the origin of the “factor”
14
responsible for lifting the 3-week SC growth barrier in mouse DRGN/SC cocultures, the
latter problem has proved to be more difficult to resolve. Therefore, we set out to
determine the conditions that would allow a faster mouse SC proliferation rate, so as to
obtain as high yields of myelination-competent cells as possible within the 3-week period
following the dissection of the nerves, without compromising the purity.
When the mouse sciatic nerves were submitted to the enzymatic dissociation
recommended for the rat nerves (Kleitman et al. 1997), the SCs emerging from the
digested nerve fragments proliferated very slowly, if they survived at all. This problem was
resolved by empirically reducing by 50 % the concentration of the trypsin/collagenase
mixture used to dissociate the nerve tissue, and by replacing the foetal calf serum in all
culture media with 10 % heat-inactivated horse serum; in preliminary assays, the horse
serum appeared to provide a better SC proliferation rate than the calf serum.
Fibroblasts present in the sciatic nerve connective tissue proliferate well in culture and can
thus rapidly take over the SC cultures if they are not eliminated soon after the culture is
started. In addition to their faster proliferation rate, fibroblasts have also been shown to
have an inhibitory effect on SC growth (Wood 1976). In the original protocols described by
Brockes et al. (1979) and Kleitman et al. (1997) for the rat, fibroblasts are eliminated by a
dual treatment with cytosine arabinoside (ARA-C), a powerful anti-mitotic agent, followed
by complement-mediated cytolysis using an anti-Thy-1.1 antibody; the Thy-1.1 antigen is
expressed on rat fibroblasts, but not on the SCs, thus providing a convenient and effective
SC selection parameter.
The Thy-1.1 antigen is not present on mouse fibroblasts, which are thus insensitive to the
complement-mediated immunocytolysis protocol using anti-Thy-1.1 antibodies. On the
other hand, mouse fibroblasts, but not mouse SCs (White et al. 1983), do express the Thy1.2 antigen and this was exploited by Dong et al. (1999), who replaced the anti-Thy-1.1
antibodies with anti-Thy-1.2 antibodies in their protocol employed to isolate SCs from
neonatal mice.
Early trials using the dual ARA-C/anti-Thy-1.2 treatment resulted, at best, in very poor SC
yields, and poor culture survival, presumably due to cytotoxic effects of ARA-C. The
treatment with ARA-C was therefore eliminated from the protocol, and the selective
elimination of the mouse fibroblasts is now performed using only one round, or
occasionally two rounds, of complement-mediated cytolysis using anti-Thy 1.2 antibodies.
A similar approach has already been successfully used by others (Seilheimer and
Schachner 1987; Kim et al. 1995; Nicholson et al. 2001).
15
Another factor that may contribute to the selective enrichment of the SCs is the use of
forskolin in all of the SC culture media already when the shredded sciatic nerves are
plated for the first time; forskolin is an adenylate cyclase activator, which has been shown
to act as a SC mitogen, while inhibiting the proliferation of fibroblasts (Brockes and Raff
1979; Raff et al. 1978; Yamada et al. 1995).
The supplementation of the culture medium with heregulin-β1 (also known as neuregulin1, or glial growth factor 2) and forskolin is sufficient to provide a satisfactory and sustained
proliferation of rat SCs in culture (Brockes et al. 1979; Kleitman et al. 1997), and others
have shown that heregulin-β1 and forskolin are sufficient to support mouse SC growth in
media with low serum contents (Nicholson et al. 2001). However, our recent results (Lahti
et al. in preparation) indicate that lower yields of mouse SCs are obtained with low serum
concentrations, and that efficient mouse SC expansion in culture requires not only the
presence of heregulin-β1 and forskolin, which can effectively ensure the survival of the
culture, but also that of bFGF and bovine pituitary extract (PE). The latter are added only
after the SC purification step, so as not to promote the growth of the fibroblasts.
Numerous studies have shown that the behaviour of cultured SCs is affected by the nature
of the substratum upon which they are cultured. Those which have been repeatedly shown
to be suitable for the proliferation of SCs of human, rat and/or murine origin include
collagen-, laminin-, and poly-L-lysine-coated plastic culture surfaces. Not surprisingly, we
also found that all three substrata supported mouse SC growth, but cells cultured on
collagen type 1 and poly-L-lysine allowed significantly higher (p < 0.05) proliferation rates
than laminin. Although the average mouse SC yield was slightly higher on collagen type 1
than on poly-L-lysine, the difference was not significant. On the other hand, rat SCs have
been shown (Porter et al. 1986) to grow better on poly-L-lysine than on ammoniated
collagen type 1, and laminin proved to be the best surface for human SC proliferation,
followed by collagen and poly-L-lysine (Casella et al. 1996).
Fibronectin-coated substrata have previously been shown to stimulate rat SC proliferation
(Baron-Van Evercooren et al. 1982). In preliminary experiments (not shown), we found that
fibronectin did indeed provide for a good mouse SC proliferative activity, but the growth
was not superior to that observed using poly-L-lysine, which costs approximately 10 times
less. Since mouse SC growth rates on poly-L-lysine and collagen type 1 were very similar,
and since poly-L-lysine coating is both faster and more economical than collagen type 1
coating, we prefer to use poly-L-lysine-coated culture dishes for the expansion of mouse
SCs.
16
The variations of the growth rate observed in the proliferation assays reported in Table 1
do not appear to correlate with the time for which the cells were actually cultured, nor with
the number of cells plated on each dish, nor with the cell plating density. On the other
hand, the proliferation rates observed in these assays were generally higher in the smaller
culture dishes. Since all of the plates were coated with poly-L-lysine, following the same
protocol, it is likely that apparent dish size-related growth rate variations result not from the
dishes themselves, but rather from factors associated with the difference between the
volumes of culture medium employed.
Indeed, in the two sets of assays performed (at the same time) using the cells from
isolation C (see Table 1a), the cells themselves, the SC Growth Medium batch, the cell
plating densities and the culture times were all identical. Despite this, the average
proliferation rate (Table 1b) was about 30% higher in the 35-mm dishes (GW = 4.30 ± 0.62)
than in the 60-mm dishes (GW = 3.31 ± 0.40). The only differences between the two culture
sets were that the culture medium volume in the 60-mm dishes was two-fold greater, while
the culture dish surface area and the number of cells seeded/dish were both 3-fold higher.
Thus, there is about 30% less medium available per cell in the larger dishes. This may
translate into a faster depletion of nutrients and/or growth-stimulating factors provided in
the culture medium, and/or a faster rate of increase in the concentration of growth
inhibitory factors secreted by the mouse SCs in these cultures.
Unlike rat SCs, which can be readily cultured through several passages and for long
periods of time, growth slows considerably after 3 weeks in pure mouse SC cultures,
irrespective of the number of passages performed during this time, and usually ceases
within 4 weeks, when the cells become quiescent. Comparable observations have been
made earlier by White et al. (1983). In our experience, subsequent passaging does not
stimulate these “quiescent” cells to re-enter a growth cycle (not shown). Under the SC
culture conditions described here, the culture medium is renewed totally every two days,
thereby removing any soluble, extracellular growth inhibitory and stimulatory factors that
may be secreted by the SCs themselves. This suggests that the causal factor for the
inhibition of the mouse SC proliferation is cumulative, either at the SC surface, or within
the cells, and must be actively removed, or antagonized, in order to re-promote
proliferation in growth-arrested cultures.
Mouse Schwann cell – neuron interactions in culture
The removal, or reversal, of the mouse SC growth inhibition may be mediated by neurons.
Evidence for this is provided by our phase-contrast microscopy observations (see Figure
17
5) that show a clear increase in SC number in the DRGN/SC cocultures, but not in cultures
of SCs without neurons performed under otherwise identical conditions. Since the culture
conditions were identical, we can rule out the possibility that factors present in the
Differentiation Medium itself are responsible for the increased SC number in the DRGN/SC
cocultures. The increase is also not the result of SCs, or SC precursors, growing out of the
DRGs, as demonstrated when the latter are cultured alone (Figures 5A and 5B), under
conditions identical to those employed in the SC cultures and DRGN/SC cocultures.
Glial growth factor (GGF), also known as neuregulin-1, or heregulin-1 in humans, is
produced by axons and, among numerous other effects, has a mitogenic activity towards
SCs (reviewed by Corfas et al. 2004). GGF would thus be a tempting candidate to explain
the growth stimulatory effect of the neurons on the SCs in the cocultures. Indeed,
exogenous heregulin-1 is absent from the Differentiation Medium employed in this
coculture assay, which could explain the lack of growth of the SCs in the absence of
neurons. However, heregulin-1 is routinely included in our SC growth medium, and is
therefore present when the SCs stop proliferating in the pure cultures. Consequently, if
neuronal GGF does play a role in the growth stimulatory effect of the DRGNs on the
mouse SCs, it is likely that the presence of some other factor(s) of neuronal origin is also
necessary for it to do so. In earlier studies using rat DRGN/SC cocultures, it was shown
that neurons and membrane fractions prepared from sensory neurons can induce the
proliferation of SCs (Salzer and Bunge 1980), and that this phenomenon requires direct
contact of the SC with the neuronal membrane (Salzer et al. 1980). It was also shown that
the (heparin-binding) mitogenic activity is not an intrinsic membrane component, but is
peripherally associated with the neuronal membrane (Ratner et al. 1988).
The mouse SCs isolated using the protocol described here maintain their ability to undergo
terminal differentiation to a myelinating phenotype and, more importantly, to form compact
myelin in culture. This was demonstrated by analysing the expression patterns of both
early (MAG) and late (MBP) markers of myelination in a three-dimensional DRGN/SC
coculture model (Päiväläinen et al. submitted), under myelination-permissive conditions.
The segregation of MBP into MAG-free regions of the sheath indicated the formation of
compact myelin, while the segregation of the MAG into small, MBP-free band-like areas
along the internodes indicated the formation of Schmidt-Lanterman incisures.
To our knowledge, this is the first time that SCs purified and expanded from neonatal
mouse peripheral nerves have been shown to be capable of forming “mature” myelin when
cocultured with embryonic mouse DRGNs. This introduces the possibility of exploiting the
18
numerous spontaneous or bio-engineered mutant mouse strains presenting PNS disorders
by performing mixed phenotype/genotype DRGN/SC cocultures. These will allow to
analyse the cumulative effects of different mutations affecting functions related to
neuronal-SC interactions, without actually having to cross-breed the animals. They will
also allow to test whether a given myelin phenotype can be reproduced through the
expression of a mutated gene by the neurons, the SCs, or by both.
19
References
Baron-Van Evercooren A., Kleinman H.K., Seppa H.E., Rentier B. and Dubois-Dalcq M.
(1982) Fibronectin promotes rat Schwann cell growth and motility. J. Cell Biol. 93:211-216.
Brockes J.P., Fields K.L. and Raff M.C. (1979) Studies on cultured rat Schwann cells. I.
establishment of purified populations from cultures of peripheral nerve. Brain Res.
165:105-118.
Brockes J.P. and Raff M.C. (1979) Studies on cultured rat Schwann cells. II. comparison
with a rat Schwann cell line. In Vitro 15:772-778.
Casella G.T., Bunge R.P. and Wood P.M. (1996) Improved method for harvesting human
Schwann cells from mature peripheral nerve and expansion in vitro. Glia 17:327-338.
Corfas G., Velardez M.O., Ko C.P., Ratner N. and Peles E. (2004) Mechanisms and roles
of axon-Schwann cell interactions. J. Neurosci. 24:9250-9260.
Dong Z., Sinanan A., Parkinson D., Parmantier E., Mirsky R. and Jessen K.R. (1999)
Schwann cell development in embryonic mouse nerves. J. Neurosci. Res. 56:334-348.
Garbay B., Fournier M., Sallafranque M.L., Muller S., Boiron F., Heape A., Cassagne C.
and Bonnet J. (1988) Po, MBP, histone, and DNA levels in sciatic nerve. postnatal
accumulation studies in normal and trembler mice. Neurochem. Pathol. 8:91-107.
Hyden H. and McEwen B. (1966) A glial protein specific for the nervous system. Proc.
Natl. Acad. Sci. U.S.A. 55:354-358.
Jessen K.R., Thorpe R. and Mirsky R. (1984) Molecular identity, distribution and
heterogeneity of glial fibrillary acidic protein: An immunoblotting and immunohistochemical
study of Schwann cells, satellite cells, enteric glia and astrocytes. J. Neurocytol. 13:187200.
Jessen K.R., Morgan L., Stewart H.J. and Mirsky R. (1990) Three markers of adult nonmyelin-forming Schwann cells, 217c(ran-1), A5E3 and GFAP: Development and regulation
by neuron- Schwann cell interactions. Development 109:91-103.
Kim H.A., Rosenbaum T., Marchionni M.A., Ratner N. and De Clue J.E. (1995) Schwann
cells from neurofibromin-deficient mice exhibit activation of p21ras, inhibition of cell
proliferation and morphological changes. Oncogene 11:325-335.
Kim H.A., Ling B. and Ratner N. (1997) Nf1-deficient mouse Schwann cells are angiogenic
and invasive and can be induced to hyperproliferate: Reversion of some phenotypes by an
inhibitor of farnesyl protein transferase. Mol. Cell. Biol. 17:862-872.
20
Kleitman N., Wood P.M. and Bunge R.P. (1997) Tissue culture methods for the study of
myelination. In: Molecular and cellular approaches to neural development (Cowan W.M.,
Jessell T.M., Zipursky S.L., Eds.), pp337-377. Oxford University Press (New York/Oxford).
Lahti O., Honkanen H., Nissinen M., Myllylä R., Päiväläinen S., Pekkanen M., Peltonen S.,
Peltonen J. and Heape A.M. (manuscript in preparation) Effects of growth factors on
mouse Schwann cell proliferation and differentiation in culture.
Lepore G., Zedda M., Gadau S. and Farina V. (2003) Purified Schwann cell primary
cultures from mouse and rat sciatic nerve. Arch. Ital. Biol. 141:207-210.
Liu N., Varma S., Shooter E.M. and Tolwani R.J. (2005) Enhancement of Schwann cell
myelin formation by K252a in the trembler-J mouse dorsal root ganglion explant culture. J.
Neurosci. Res. 79:310-317.
Manent J., Oguievetskaia K., Bayer J., Ratner N. and Giovannini M. (2003) Magnetic cell
sorting for enriching Schwann cells from adult mouse peripheral nerves. J. Neurosci.
Methods. 123:167-173.
Nicholson S.M., Gomes D., de Nechaud B. and Bruzzone R. (2001) Altered gene
expression in Schwann cells of connexin32 knockout animals. J. Neurosci. Res. 66:23-36.
Nobbio L., Gherardi G., Vigo T., Passalacqua M., Melloni E., Abbruzzese M., Mancardi G.,
Nave K.A. and Schenone A. (2006) Axonal damage and demyelination in long-term dorsal
root ganglia cultures from a rat model of Charcot-Marie-Tooth type 1A disease. Eur. J.
Neurosci. 23:1445-1452.
Päiväläinen S., Nissinen M., Honkanen H., Lahti O., Peltonen J. Peltonen S. and Heape
A.M. (submitted for publication) Myelination in mouse dorsal root ganglion/Schwann cell
cocultures.
Pannunzio M.E., Jou I.M., Long A., Wind T.C., Beck G. and Balian G. (2005) A new
method of selecting Schwann cells from adult mouse sciatic nerve. J. Neurosci. Methods
149:74-81.
Porter S., Clark M.B., Glaser L. and Bunge R.P. (1986) Schwann cells stimulated to
proliferate in the absence of neurons retain full functional capability. J. Neurosci. 6:30703078.
Raff M.C., Hornby-Smith A. and Brockes J.P. (1978) Cyclic AMP as a mitogenic signal for
cultured rat Schwann cells. Nature 273:672-673.
21
Ratner N., Hong D.M., Lieberman M.A., Bunge R.P. and Glaser L. (1988) The neuronal
cell-surface molecule mitogenic for Schwann cells is a heparin-binding protein. Proc. Natl.
Acad. Sci. U.S.A. 85:6992-6996.
Salzer J.L. and Bunge R.P. (1980) Studies of Schwann cell proliferation. I. an analysis in
tissue culture of proliferation during development, wallerian degeneration, and direct injury.
J. Cell Biol. 84:739-752.
Salzer J.L., Williams A.K., Glaser L. and Bunge R.P. (1980) Studies of Schwann cell
proliferation. II. characterization of the stimulation and specificity of the response to a
neurite membrane fraction. J. Cell Biol. 84:753-766.
Seilheimer B. and Schachner M. (1987) Regulation of neural cell adhesion molecule
expression on cultured mouse Schwann cells by nerve growth factor. EMBO J. 6:16111616.
Seilheimer B., Persohn E. and Schachner M. (1989) Antibodies to the L1 adhesion
molecule inhibit Schwann cell ensheathment of neurons in vitro. J. Cell Biol. 109:30953103.
Sereda M., Griffiths I., Puhlhofer A., Stewart H., Rossner M.J., Zimmerman F., Magyar
J.P., Schneider A., Hund E., Meinck H.M., Suter U. and Nave K.A. (1996) A transgenic rat
model of Charcot-Marie-Tooth disease. Neuron 16:1049-1060.
Verdu E., Rodriguez F.J., Gudino-Cabrera G., Nieto-Sampedro M. and Navarro X. (2000)
Expansion of adult Schwann cells from mouse predegenerated peripheral nerves. J.
Neurosci. Methods 99:111-117.
White F.V., Ceccarini C., Georgieff I., Matthieu J.-M., Costantino-Ceccarini E. (1983)
Growth properties and biochemical characterization of mouse Schwann cells cultured in
vitro. Exp. Cell Res. 148:183-194.
Wood P.M. (1976) Separation of functional Schwann cells and neurons from normal
peripheral nerve tissue. Brain Res. 115:361-375.
Yamada H., Komiyama A. and Suzuki K. (1995) Schwann cell responses to forskolin and
cyclic AMP analogues: Comparative study of mouse and rat Schwann cells. Brain Res.
681:97-104.
22
Table 1a:
Variable parameters in the Schwann cell (SC) Proliferation Assays
Proliferation Assay Conditions
Isolation (a, b)
Culture dish
diameter (c)
Number of SCs
seeded/dish
SC plating density (d)
Duration of culture
(days)
(mm)
(NS)
(cells/mm2)
(T)
A
60 [n = 5]
190 000
67
4
B
60 [n = 3]
135 000
48
8
60 [n = 3]
150 000
52
7
35 [n = 4]
50 000
D
35 [n = 3]
60 000
62
8
E
35 [n = 3]
60 000
62
8
C
Table 1b:
Culture dish
diameter
Results of SC Proliferation Assays
Fold growth/week
Total yield
SC number at 1st passage
GW
NTOT = N1 x GW
N1
(m ± SD)
(m ± SD) x106
A
990 000
2.86 ± 0.69
2.83 ± 0.68
B
724 000
2.17 ± 0.59
1.57 ± 0.43
C
770 000
3.31 ± 0.40
2.55 ± 0.31
Isolation
(mm)
60
Overall
2.72 ± 0.77 [n = 11]
C
770 000
4.30 ± 0.62
3.31 ± 0.48
D
700 000
2.69 ± 0.25
1.88 ± 0.18
E
820 000
4.03 ± 0.49
3.30 ± 0.40
35
Overall
3.74 ± 0.85 [n = 10]
23
Legend to Table 1
[n]
N1
Ns
T
GW
NTOT
Number of assay cultures performed.
The total Schwann cell yield obtained at the 1st passage (i.e. following the
complement-mediated cytolysis step).
The number of cells seeded for each proliferation assay culture.
The duration (in days) of the proliferation assay cultures.
The fold growth/week (i.e. the relative increase of the Schwann cell population in
each assay normalized to one week in culture) was calculated for each culture,
using the formula:
GW = N2/Ns x T/7, where N2 is the number of cells harvested from the culture
at the second passage (i.e. the end of the proliferation
assay)
The mean [m] GW (± standard deviation [SD]) was determined for each set of
culture conditions.
Overall GW values (m ± SD) were calculated from the GW values of all the individual
assay cultures grown on dishes of the same size, irrespective of the other
parameters.
Total yields expected after culturing for 1 week following the 1st passage were
calculated using the formula:
NTOT = GW x N1,
The mean NTOT (± SD) was determined for each set of culture conditions.
Notes:
(a)
(b)
Schwann cells were isolated and cultured as described in Materials and Methods,
using the variable parameters defined in the Table 1a.
Schwann cells from isolations D & E (1st passage) were also employed for the
Growth Surface (poly-L-lysine, collagen, laminin) comparison assays reported in
Figure 1, where they are referred to as isolations I & II, respectively. Only the
results for the poly-L-lysine-coated dishes are included in Table 1b.
(c)
SC Growth Medium volumes are 1 mL (35-mm Ø dishes) and 2 mL (60-mm Ø
dishes).
(d)
Culture dish surface areas are 962 mm2 (35-mm Ø dishes) and 2827 mm2 (60-mm
Ø dishes).
24
Table 2.
Summary of principal differences between the protocols for the isolation,
purification and expansion of rat and mouse SCs.
Step
Enzymatic
dissociation of
Rat Schwann cells (1,2)
Mouse Schwann cells
0.25 % trypsin
0.125 % trypsin
0.1 % collagenase A
0.05 % collagenase A
ARA-C anti-mitotic treatment
Complement-mediated cytolysis
followed by complement-
of fibroblasts using anti-Thy 1.2.
sciatic nerves
Schwann cell
purification
mediated cytolysis of fibroblasts
antibody
(3)
Effect of
change
Improved
cell yield
Increased
SC survival
using anti-Thy 1.1 antibody
HG-DMEM containing
HG-DMEM containing (4)
10 % foetal bovine serum
10 % inactivated horse serum
4 mM L-glutamine
4 mM L-glutamine
Schwann cell growth
100 u/mL penicillin/streptomycin
100 u/ml penicillin/streptomycin
Increased
(medium composition)
2 ng/mL human heregulin-β1
2 ng/ml human heregulin-β1
SC growth
4 μM forskolin
0.5 μM forskolin
10 ng/ml human bFGF
20 μg/ml bovine pituitary extract
Notes
(1)
Brockes et al. 1979.
Kleitman et al. 1997.
(3) Dong et al. 1999.
(4) Lahti et al. In preparation.
(2)
25
Legends to Figures
Figure 1
Growth of SCs on different coat surfaces. Triplicate cultures (1-3) of SCs from two
separate isolations (I & II) were grown on each of the three substrata (poly-L-lysine [PLL],
collagen type 1 [COL] and laminin [LAM]) in the presence of heregulin-β1, forskolin, FGFb
and PE, as described in Materials and Methods. After 8 days, the cells in each culture
were harvested and counted. Each bar represents one culture, and the means (± SD) for
the triplicate cultures are given above each set. The horizontal dashed line indicates the
number of SCs seeded onto each dish (6 x 104 cells/dish) at the start of the experiment.
Figure 2.
Phase-contrast microscopy of the SC cultures assayed in Figure 1 after 8 days on A) polyL-lysine, B) collagen, and C) laminin. A 20x objective and the fields were enlarged with the
optical zoom of the camera.
Figure 3.
Characterization of purified SC cultures by fluorescence microscopy. After one round of
complement-mediated cytolysis, the cultures were triple-labelled with Hoechst nuclear
stain (blue), and antibodies to GFAP (green) and S100 (red). Images A-D show a typical
field observed in these cultures.
A) Hoechst nuclear stain; B) anti-GFAP antibody; C) anti-S100 antibody; D) merged
image formed by superimposing the images A-C. Objective 40x.
Figure 4.
Phase-contrast microscopy of cultures at different phases of the mouse SC isolation
protocol. A) Two days after plating, SCs (black double arrow-heads) emerge from
shredded nerve fragments (white arrows); B) SC culture in Basic Growth Medium, 8 days
after isolation from the sciatic nerves, following complement-mediated cytolysis of
fibroblasts, and prior to the first passage; C) purified SC culture in SC Growth Medium, 5
days after the first passage (total of 13 days in culture). Objective 4x (A) and 10x (B & C).
Figure 5
Neuron-mediated reactivation of mouse SC growth. Pre-purified DRGNs (A, B, E, F; one
DRG/culture) and SCs (C, D, E, F; 15 x 103 cells/culture) were cultured alone (A-D) and
together in cocultures (E, F) under otherwise identical conditions, as described for the 3-
26
dimensional DRGN/SC cocultures (see Materials & Methods). Phase contrast microscope
images were taken of representative cultures 2 days after seeding the SCs onto the
Matrigel in DRG Growth Medium (A, C, E), and again after 7 days culture in Differentiation
Medium (B, D, F). Note the absence of SCs in the DRGN cultures (A, B), the similar cell
numbers in the SC cultures (C, D), and the significant increase in the number of SCs when
the latter are cocultured with DRGNs (E, F). The inset (f’) is a digital enlargement of the
boxed area in F, illustrating the high SC density after 7 days. White arrows point to DRGs,
black double arrowheads point to neuronal processes, and black arrows point to SCs.
Figure 6.
Myelination in cocultures of pre-purified mouse SCs and DRGNs. Three-dimensional
myelinating mouse DRGN/SC cocultures were performed in 96-well culture plates as
described in the Materials & Methods section. After 3 weeks in the Myelination Medium,
the cocultures were fixed, immunolabelled with mouse monoclonal anti-MAG antibody
(clone 513) and rabbit polyclonal anti-MBP antibody, and examined by confocal
microscopy, as described in Materials & Methods. Phase contrast microscopy was
performed after 3½ weeks in Myelination Medium.
The immunofluorescence labelling patterns of MBP (A, a) and MAG (B, b) after 3 weeks in
Myelination Medium are typical of mature myelin structures, in which MAG and MBP are
well segrated from each other (C, c). Note the band-like expression of MAG along the
MBP-positive internodes, indicating the formation of Schmidt-Lantermann incisures (white
arrows) in the compacted sheaths (yellow arrows).
Scale bars in A, B & C: 20 m. Images a, b and c are digital enlargements of the same
boxed area in images A, B and C, respectively, and were acquired using the software of
the confocal microscope.
(D) Phase contrast microscopy reveals abundant figures indicative of myelin formation in
myelinating mouse DRGN/SC cocultures after 3½ weeks. Double-headed arrows point to
4 of the many (~25) myelin sheaths visible in this field.