As published in:
Acta Chururgica Austriaca 30 (Supplement 147): 15-19, 1998

Successfully culturing Schwann cells from adult peripheral nerve

A.D. Ansselin*, S.D. Corbeil & D.F. Davey

Electron Microscope Unit (F09)*,
Department of Physiology (F-13),
and Institute for Biomedical Research,
University of Sydney
NSW 2006


There are numerous reasons why culturing adult Schwann cells is essential: to study the membrane properties of Schwann cells and axon-Schwann cell communication and how these might be altered in neuropathic conditions; to use Schwann cells for the repair of lesioned peripheral nerve; or to exploit their potential for regeneration in CNS lesions.

Much of what we know about Schwann cell properties has come from studies using cultured embryonic or neonatal cells. We now have evidence that adult Schwann cells differ from neonatal cells in the expression of some receptors(1), and the differences may well extend into other functions of Schwann cells. The use of Schwann cells in surgical repair of lesioned peripheral nerves or CNS lesions also necessitates the use of adult cells derived from the recipient, since allogeneic Schwann cells are subject to rejection(2,3). It is therefore important that adult Schwann cells can be isolated and cultured, and that we have the ability to expand these cultures and harvest large numbers of cells.

Unfortunately, Schwann cells are difficult to isolate from adult mammalian peripheral nerves because of the abundance of connective tissue and the highly differentiated state of the cells, particularly those involved in the formation of myelin. Few cells are successfully cells isolated from fresh nerve. They do not attach readily to a substratum, are refractory to proliferation, and many die within the first 24h after plating.

It has been shown that in vivo, both neurons and Schwann cells are conditioned by a nerve lesion, speeding up Schwann cell proliferation and neuronal regeneration; the effect being most pronounced in the first three days after applying the lesion(4). By triggering Wallerian degeneration, Schwann cells retract from the axons they ensheath, de-differentiate and enter the mitotic cycle. The cells are then primed for the regenerative process, and are easier to culture. Observations on Schwann cells in previous regenerative studies(5,6) suggested that some time during the first two weeks following a lesion might be optimal to maximise the yield of conditioned Schwann cells from the nerve distal to the lesion.

In this study on adult rat peripheral nerve, we report how a conditioning lesion improves both the yield of Schwann cells and the number of cells which successfully attach to a tissue culture substratum. We also show that a much less aggressive isolation procedure can be adopted, improving the yield further. We further discuss how a modification of this technique has now allowed us to successfully isolate and culture adult cells from human peripheral nerve, thereby demonstrating the clinical application of the technique.

Materials and Methods

All operative procedures and post-operative care of the experimental animals were carried out according to the guidelines set out by the National Health and Medical Research Council of Australia.

Conditioning lesion: Adult rats (4-5 months old) were anaesthetised using 1.8 % halothane in O2. The left sciatic nerve was exposed, severed at the sciatic notch and deflected. At the appropriate survival time, a 20 mm segment of nerve was excised from the distal stump of the conditioned sciatic nerve and the unoperated (control) side, using sterile techniques.

Preparation of cultures: The excised nerve segments were:

  1. Transferred to sterile Petri dishes containing Ca2+ and Mg2+ free Dulbecco's phosphate buffered saline (CMF);
  2. gently stripped of the epineurium;
  3. minced in CMF, transferred to a vial and centrifuged for 5 min at 1000rpm.
  4. The supernatant was replaced with 10 ml RPMI-1640 medium containing 1.25 units/ml Dispase, 0.05% (w/v) collagenase, and 0.1% (w/v) hyaluronidase.
  5. The pellet was gently triturated, then either incubated for 3-4 h or overnight at 37°C. With the short incubation batches, the suspension was gently agitated hourly to determine the state of dissociation.
  6. The cell suspension was centrifuged, rinsed in CMF, centrifuged and resuspended in 1 ml of RPMI-1640 (ICN) supplemented with:
  7. A sample was used to determine cell density, while the remainder were plated (100µl aliquots) onto laminin-coated glass coverslips in 24-well plates at a density of 2.5 × 106 cells per well. The cell suspension was allowed to settle for 20-30 min before the wells were carefully filled with complete medium. The cells were left to attach for 24 h, rinsed with CMF and new medium added. Medium was replaced twice a week.

Counting of attached cells: Only cells displaying the typical bipolar or tripolar morphology, and therefore estimated to be truly attached, were counted. Twenty-four hours after plating, the total number of attached Schwann cells was estimated for each nerve segment using a random sampling technique. The sampled area averaged 3% of the culture well/plate. The count for all the wells used for one nerve segment was averaged to give one datum per animal.

The identity of the cells was verified with antibody staining using S-100 (Sigma) or Ran-1 antiserum (kindly donated by Dr. J. de Vellis, UCLA).

Antibody staining: The cultures were fixed (10 min, 4% paraformaldehyde in PBS), rinsed in PBS for 10 minutes, and incubated in primary antibody for 3h at 4°C. The cultures were washed in PBS, incubated in biotinylated secondary antibody for 1h at room temperature, washed in PBS (2×10 minutes), incubated in streptavidin/FITC for 1h at room temperature and rinsed with PBS. The coverslips were mounted on glass slides, observed using fluorescence optics, and photographed.


The lesion caused a progressive and significant increase in the number of cells obtained as early as 48 h after plating (P = 0.007), until day 12 (Fig. 1). Thereafter, the number of cells isolated declined significantly (P <= 0.01), although it remained above control values (P < 0.001). The different digestion times had no noticeable effect on the number of cells isolated from either conditioned or control nerves.

Fig. 1: Number of cells (mean ± s.e.) isolated from a 20 mm length of rat sciatic nerve prior to plating. Conditioned nerves were lesioned 1 - 14 days prior to the dissociation procedure. Unconditioned nerves were the control (unoperated) sciatic nerve of the experimental animals.

At all times, even as early as 24h post-lesion, more of the Schwann cells isolated from the lesioned nerve attached to the culture substratum after plating than did those from control (unlesioned) nerve (P < 0.001; Fig. 2). Decreasing the duration of enzyme digestion to between 3 and 4h markedly increased the number of attached cells without reducing the number of cells isolated. There was a small increase in the number of attached cells between day 1 and day 3, followed by a gradual but consistent increase over the following 8 days. Peak numbers of attached Schwann cells were observed between day 12 and day 14 post-lesion (Fig. 2). This means that although the number of cells isolated from the lesioned nerves had already declined significantly at day 13 post-lesion (refer to Fig. 1; P = 0.01), the cells still attached well to the substratum.

Fig. 2: Number of Schwann cells (mean ± s.e.) attached to the substratum 24 h after plating. Three relations are shown: Controls (closed circles) and for lesioned nerve, a short (triangles) or long digestion (open circles) time during the isolation procedure. Conditioning the sciatic nerve significantly increased the number of Schwann cells which successfully attached to the substratum (P < 0.001) from 24 h post-lesion onwards. Peak numbers of attached cells were found between day 12 and day 14 inclusive. Note that variability for the day 14 data is high and is not significantly different from the day 12-13 data. The two techniques used to isolate Schwann cells from lesioned nerves show a significant difference from day 4 onwards (P <=0.001), with more cells isolated using the shorter digestion time attached to the substratum.

Fig. 3 expresses the data shown in Fig. 2 as the percentage of cells which attached successfully after isolation. The greatest percentage of attached Schwann cells were found between days 12 and 14 post-lesion. It can be noted however that the standard error for the day 14 counts is large, indicating a substantial variability in the number of cells attaching to the substratum. In the majority of preparations (60%), more than 6% of the cells plated attached to the substratum. In the remainder, the number of attached cells was down, suggesting the peak period had been passed.

Fig. 3: Number of attached Schwann cells expressed as a percentage of the number of cells isolated (mean ± s.e.). The best and most consistent results were obtained 12-13 post-lesion. At day 14, the mean number of isolated cells which attached to the substratum was higher but very variable and not significantly different from day 13.

Identification of cells: Most of the attached Schwann cells had the typical oval-shaped cell body, with a prominent nucleus and bipolar extensions, giving an overall spindle shape. The identity of the cells counted was confirmed as Schwann cells with S-100 and Ran-1 antibody staining (Fig. 4). Although bipolar cells were most commonly found, some tripolar cells were found which also labelled with S-100 antibody (see Fig. 4D). Labelled round cells were not counted, being judged as not being attached. These cells disappeared when the medium was changed.

Fig. 4: Phase (A) and fluorescent (C) micrographs of confluent cultures and individual (B & D) adult Schwann cells labelled with antibody against S-100. Note the alignment of the cells in the confluent cultures, most of which showed the bipolar morphology more clearly seen in C. Occasionally, cells displayed a tripolar morphology, but this was mainly observed in areas where the cells were more sparsely distributed. Scale bar represents 50µm.


The method outlined here shows that more Schwann cells can be isolated from adult peripheral nerve having undergone Wallerian degeneration for 10-12 days than from control nerve (Fig. 1). By initiating Wallerian degeneration with the conditioning lesion, the Schwann cells have had time to separate from the axons, shed the myelin where necessary, and proliferate. The increased cell yield observed, supports this hypothesis.

The conditioning also allowed the Schwann cells to separate from the basal lamina and surrounding connective tissue, becoming mobile. The connective tissue itself seemed easier to break down. Since the cells were more easily isolated from the nerve tissue, it was possible to use milder enzymes over a relatively short digestive period with no loss in the number of cells isolated. The milder isolation technique correlated with a large increase in the number of attached cells (see Fig. 2). This suggests that enzyme digestion not only breaks down the connective tissue and extracellular matrix, but damages the cells in some way, the plasma membrane being the structure most likely damaged. Minimising enzymatic damage to the plasma membrane is crucial if the physiological status of the Schwann cells is to be studied in vitro. It is also of considerable importance if cultured adult Schwann cells are to be studied for diagnostic purposes.

The greater variability observed at day 14 suggests that the Schwann cells are changing in some way. Fewer cells were isolated, and these no longer reliably attached to the substratum. Perhaps the initial stimulus provided by the degenerating axons had declined. Since the lesioned nerve was not reinnervated, there would also be a lack of any stimulating factor from regenerating sprouts. The fact that at day 21 (data not shown) both the number of Schwann cells isolated and the number attached to the substratum had fallen to control levels supports this observation.

Although the technique described in this study gives much improved yields of attached and viable adult Schwann cells, the success rate is still small when the total number of cells available and plated is considered (see Fig. 3). The counting of cells prior to plating is a standard procedure, but at best it is only a guide to the maximum number of cells isolated. It is likely that the number of one specific cell type, in this case the Schwann cells, is overestimated since it is difficult to absolutely differentiate Schwann cells from other non-neuronal cells, for example fibroblasts, at that stage of isolation. There is, however, another more serious problem to be considered. It is possible that what is still a reasonably aggressive isolation technique selects a sub-population of Schwann cells. In this case, the cells which are successfully cultured would not be a true representation of the in vivo population. The fact that our cultured cells were successfully used in transplants (see following paper(7)) would tend to argue against the technique selecting non-myelinating Schwann cells in vivo. Other studies have also confirmed that cultured Schwann cells can form myelin in vitro (8-10), but these studies used embryonic or neonatal Schwann cells. However, there is some evidence that non-myelinating Schwann cells proliferate very rapidly during Wallerian degeneration(11). It is therefore possible that non-myelinating Schwann cells are overly represented in a culture system. There may also be more subtle differences related to the ability of cells to survive the isolation technique depending on their pre-isolation health status, an important factor in obtaining cells from diseased peripheral nerve. It is therefore important that further studies be carried out to establish if there are sub-populations of Schwann cells.

Finally, the advantage of this technique over previous reports(12-14) lies in the improved yields associated with the considerably shortened digestion phase. This reduces the time taken to isolate the cells, results in more cells attaching to the substratum and the cells are in a proliferative state reaching confluence within 5-7 days.

It is obvious that for clinical applications, conditioning the nerve in vivo while optimal for cell yield, would not be the preferred option. We have subsequently modified and refined the technique to allow us to obtain confluent cultures of adult human Schwann cells. The source of nerve came from 3 mm segments of sural nerve obtained as part of diagnostic procedures or phrenic nerve obtained from autopsy material.

As with the animal Schwann cells, the critical factor in successfully isolating and culturing the adult human Schwann cells, is allowing the cells to separate themselves from the axons they ensheath, de-differentiate and become mobile. We found that by cleaning the nerve specimens of connective tissue, teasing the fibres apart gently, and chopping the pieces up into small pieces (1mm3), we can maintain the pieces in culture in complete medium for 5-7 days to allow Wallerian degeneration to proceed. After this time, the nerve segments can be processed following the protocol detailed in the Methods with one modification. Human nerve has a lot more connective tissue, and we find that enzyme digestion takes 5-6h. The digestion phase is monitored hourly, by gently agitating the suspension and checking how the connective tissue is disintegrating. We do not allow the digestion phase to go beyond 6h, preferring to plate small clumps if not completely dissociated. The cells will migrate out of the clumps. Extending beyond that time results in a reduced yield of attached cells. We have found that a 3mm biopsy yields enough cells to plate in one well of a 12-well plate. Having such small biopsies, we have not been able to sacrifice samples to estimate the number of cells isolated accurately, it is estimated to be in the range of 1×106 per mm of nerve. In our hands, the cells from the biopsies reached confluence in 7-10days, but the results were very variable. Cells from autopsied material were much more difficult to culture. We believe that to be mostly related to the delay in time since the death of the donor (usually 12-24h). However, there is a possibility that Schwann cells from biopsied nerves were pre-conditioned since most (but not all) nerves were neuropathic. We need to explore this possibility further. The in vitro technique has proved to be so successful we now only use this technique to culture all adult Schwann cells from a wide range of peripheral nerves including rats and mice, rabbits, guinea pigs and non-human primates as well as human tissue.


1. Ansselin AD, Davey DF, Allen DG: Extracellular ATP increases intracellular calcium in cultured adult Schwann cells. Neuroscience (Oxford) 1997;76:947-955.
2. Ansselin AD, Pollard JD: Immunopathological factors in peripheral nerve allograft rejection: quantification of lymphocyte invasion and major histocompatibility complex expression. J Neurol Sci 1990;96:75-88.
3. Ansselin AD, Westland K, Pollard JD: Low dose, short term Cyclosporin A does not protect the Schwann cells of allogeneic nerve grafts. Neurosci Lett 1990;119:219-222.
4. Sjöberg J, Kanje M: Effects of repetitive conditioning crush lesions on regeneration of the rat sciatic nerve. Brain Res 1990;530:167-169.
5. Ansselin AD, Davey DF: Axonal regeneration through peripheral nerve grafts: The effect of proximo-distal orientation. Microsurgery 1988;9:103-113.
6. Ansselin AD, Davey DF: The regeneration of axons through normal and reversed peripheral nerve grafts. Restor Neurol Neurosci 1993;5:225-240.
7. Ansselin AD, Fink T, Davey DF: An alternative to nerve grafts in peripheral nerve repair: Nerve guides seeded with adult Schwann cells. Acta Chururgica Austriaca 1998;30(Supplement 147):19-24.
8. Eldridge CF, Bunge MB, Bunge RP, Wood PM: Differentiation of axon-related Schwann cells in vitro. I Ascorbic acid regulates basal lamina assembly and myelin formation. J Cell Biol 1987;105:1023-1034.
9. Owens GC, Bunge RP: Schwann cells infected with a recombinant retrovirus expressing myelin-associated glycoprotein antisense RNA do not form myelin. Neuron 1991;7:565-575.
10. Wood P, Moya F, Eldridge C, Owens G, Ranscht B, Schachner M, Bunge M, Bunge R: Studies of the initiation of myelination by Schwann cells. Ann NY Acad Sci 1990;605:1-14.
11. Clemence A, Mirsky R, Jessen KR: Non-myelin-forming Schwann cells proliferate rapidly during Wallerian degeneration in the rat sciatic nerve. J Neurocytol 1989;18:185-192.
12. Komiyama A, Novicki DL, Suzuki K: Adhesion and proliferation are enhanced in vitro in Schwann cells from nerve undergoing Wallerian degeneration. J Neurosci Res 1991;29:308-318.
13. Rutkowski JL, Tennekoon GI, McGillicuddy JE: Selective culture of mitotically active human Schwann cells from adult sural nerves. Ann Neurol 1992;31:580-586.
14. Morrissey TK, Kleitman N, Bunge RP: Isolation and functional characterization of Schwann cells derived from adult peripheral nerve. J Neurosci 1991;11:2433-2442.


This work was supported by the Microsearch Foundation of Australia.

© A.D. Ansselin, S.D. Corbeil & D.F. Davey 1998
Address for correspondence:
A/Prof. D.F. Davey
Department of Physiology (F-13)
University of Sydney
NSW 2006 Australia

Phone: +61 2 9351 4559 Fax: +61 2 9351 5182 Email:

14,683 accesses.