Labeling Schwann cells with CFSE*/an in vitro and in vivo study

Xiuming Li, Hector Dancausse, Israel Grijalva, Maria Oliveira, Allan D.O. Levi *

The Miami Project to Cure Paralysis, and Department of Neurosurgery, University of Miami School of Medicine, Miami, Lois Pope LIFE Center

(R-48), 1095 NW 14th Terrace, Miami, FL 33136, USA

Received 7 October 2002; received in revised form 3 February 2003; accepted 3 February 2003

Abstract

Schwann cell (SC) transplantation is a promising strategy for axonal regeneration in the nervous system. Identifying the grafted

SCs is an important aspect of this approach. The current study sought to establish a simple, reliable, fluorescent labeling method for

SCs with a lipophilic molecule, 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFSE). Human SCs were incubated with

varying concentrations of CFSE for different time periods. Based on the viability of labeled SCs and its plating efficiency, 1 min

incubation with 5 mM CFSE at 37 8C was selected as the optimal labeling condition. Flow cytometric analysis and fluorescence

microscopy demonstrated that the fluorescence of labeled SCs would fade over 4 weeks. Immunostaining for the phenotypic

expression of SC markers, including S100, GFAP, P75, and MHC-I/II at 1 and 4 weeks after incubation with CFSE showed no

difference between labeled and non-labeled SCs. Mixed cultures of labeled human SCs and rat SCs for 48 h were performed in

triplicate and demonstrated that no leakage of dye from labeled SCs in cell culture occurred across species. A total of 14 injections of

2�/105 labeled SCs were performed within the spinal cord at T8 and/or L1 level in 9 nude rats. The animals were euthanized at 1 (6

injections) and 4 weeks (8 injections). Grafted labeled SCs survived for at least 4 weeks, and could be easily recognized in the nude

rat spinal cord without leakage of dye to surrounding cells. The SCs migrated in white and gray matter 3�/6 mm away from the

injection and in the central canal for up to 12 mm. These results suggest that CFSE can be used as a fluorescent tracer of human SCs

for both in vitro and in vivo studies, for a period of at least 4 weeks.

# 2003 Elsevier Science B.V. All rights reserved.

Keywords: CFSE; Schwann cells; Fluorescence; Labeling; Transplantation; Spinal cord; Nude rats

1. Introduction

Schwann cell (SC) transplantation is a promising

strategy in encouraging regeneration from both the

central and peripheral nervous system (Berry et al.,

1988; Felts and Smith, 1992; Levi and Bunge, 1994; Xu

et al., 1995a,b, 1997; Plant et al., 2001). However,

identification, localization and distinguishing the

grafted SCs from the host within the spinal cord and/

or peripheral nerves can be difficult. Reliable markers

are needed to label SCs to investigate the contribution of

transplanted SCs to the regeneration process. The ‘ideal’

reporter molecule should be stable, safe, should not

undergo alterations due to the adjacent environment

(e.g. no leakage), and be simple to apply.

Numerous reporter molecules have been studied for

different cell labeling and tracing techniques with

various limitations and results. Radioactive probes

such as 125Iodine and 51Chromium for lymphocyte

labeling have poor uptake, are toxic with internal

radiation effects, and have rapid elution (Horan et al.,

1990). Fluorescent probes such as fluorescein and

rhodamine isothiocyanate tend to rapidly escape or

leak (Butcher et al., 1980; Olszewski, 1987). Indocarbo-

cyanine dye results in a non-uniform labeling for

neuronal cells (Honig and Hume, 1986). The use of

chemical markers such as Hoechst 33342 (nucleus) and

PKH26 (membrane) are hindered by dye leakage, cell

toxicity, reduced phenotypic expression of the SC, and

short term expression (Horan and Slezak, 1989; Ever-

cooren et al., 1991; Casella et al., 1996; Ansselin et al.,

1997; Hermanns et al., 1997; Mosahebi et al., 2000). SCs

transduced with recombinant gene markers such as

retrovirus encoding the bacterial b-galactosidase gene

* Corresponding author. Tel.: �/1-305-243-2088; fax: �/1-305-243-

6017.

E-mail address: alevi@med.miami.edu (A.D.O. Levi).

Journal of Neuroscience Methods 125 (2003) 83�/91

www.elsevier.com/locate/jneumeth

0165-0270/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.

doi:10.1016/S0165-0270(03)00044-X

is an advanced labeling method but requires cell division

to introduce the gene, thus transfection efficiency

becomes an issue (Mosahebi et al., 2001), particularly

in human primary cells.5-(and-6)-carboxyfluorescein diacetate succinimidyl

ester (CFDA, SE; CFSE) is a lipophilic molecule that

is only minimally fluorescent until it is transported

inside the cells, where esterase cleaves the acetyl groups

and the molecule becomes markedly fluorescent. The

succinimidly ester group covalently binds to amino

groups on intracellular macromolecules, anchoring the

dye. CFSE has been used predominantly to observelymphocytes proliferation, migration and to track

neuronal cells (Weston et al., 1990; Paramore et al.,

1992; Hasbold et al., 1999). It has been found to be

stable and nondiffusible both in vitro and in vivo.

Furthermore, researchers have reported using CFSE

labeling of hepatocytes for localization following in-

traportal transplantation (Karrer et al., 1992; Fujioka et

al., 1994). In this study, we describe the conditions forlabeling human SCs with CFSE that demonstrate

labeling stability and safety without evidence of leakage

in vitro. Subsequently, we used this marker to localize

labeled human SCs with CFSE in the spinal cord of

nude rats in vivo. This method of labeling is simple and

nontoxic to human SCs at the concentration and

incubation time selected, and it allows in vitro identifi-

cation and in vivo localization of engrafted human SCs.

2. Materials and methods

2.1. Preparation of purified human Schwann cells

Human peripheral nerves were obtained from organ

donors by the transplant procurement team of the

University of Miami School of Medicine. The nerves(cauda equina) were harvested and processed for cell

culture as previously described (Casella et al., 1996; Levi

et al., 1996). In brief, the nerves were harvested and

washed three times in Leibovitz’s L-15 (Gibco) and

perineurial fascicles were stripped of connective tissue

and blood vessels. The fascicles were cut into 1�/2 mm

blocks and kept in suspension in D10 media (DMEM,

Dulbecco’s Modified Eagles medium, Gibco, GrandIsland, NY) supplemented with 10% fetal bovine serum

(FBS). The nerve explants were dissociated according to

the protocol of Pleasure et al. (1986). The explants were

subjected to gentle enzymatic dissociation after an

overnight incubation in an enzyme cocktail consisting

of 0.05% collagenase (Worthington Biochemicals Corp.,

NJ) and 1.25 U/ml dispase (Boehringer Mannheim

Biochemicals, Germany) and 15% FBS in DMEM.The following morning, the explants were gently tritu-

rated with a straight glass borosilicate pipette until

individual explants could no longer be recognized. After

four washes in L-15 and 10% FBS, the cell suspension

was plated on ammoniated collagen-coated tissue cul-

ture dishes with a double mitogen cocktail consisting of

a recombinant form of heregulin b-1 (10 nM) and

forskolin (1 mM). The cells were replated to new dishes

when they achieved confluence, approximately once a

week. The cells were passaged from two to six times in

the presence of mitogens prior to the in vitro and in vivo

study.

2.2. Preparation of fluorescent dye*/CFSE

Since CFSE is relatively insoluble in aqueous solu-

tions, the low-toxicity dispersing agent Pluronic F-127

(Molecular Probes, Inc., Eugene, OR) was applied to

facilitate cell loading. This nonionic detergent was used

to make up a 20% (w/v) solution in anhydrous

dimethylsulfoxide. Then, a 10 mM CFSE stock solution

was prepared by dissolving 2.8 mg CFSE (Molecular

Probes Inc., Eugene, OR) into 500 ml of this solution.

The CFSE solution must be used for labeling SCs

immediately due to its instability.

2.3. Multiple tube testing for optimal labeling of human

SCs

Human SCs were harvested and washed three times

with PBS. After measuring the viability, 5�/104 SCs in

PBS were put in each of 16 test tubes. Different amounts

of CFSE stock solution were added into triplicate tubes

to make the final concentrations of 5, 10, 20, 33 and 50

mM CFSE and one tube without CFSE as control. Three

tubes from each concentration were incubated at 37 8Cwith shaking for 1, 5 or 10 min, respectively. The SCs

were washed three times with PBS following incubation.

The viability of SCs in each of the 16 tubes was

measured by trypan blue staining. Then the SCs were

plated on collagen-coated tissue culture dishes with D10

media (DMEM, Dulbecco’s Modified Eagles Medium,

Gibco, Grand Island, NY) supplemented with 10% fetal

bovine serum (FBS) and 10 nM heregulin b1 and 1 mm

forskolin. The viability of labeled SCs in each tube was

also assessed by determining their plating efficiency,

which was the fraction of SCs attached to the dish over

the total number of the SCs in the first 24 h after plating.

The cells were replated to new dishes when they achieved

confluence, approximately every 7�/10 days. These

experiments were performed in quadruplicate. Based

on the cell viability and plating efficiency results and

statistical analysis, an optimal labeling condition was

selected to perform the following in vitro and in vivo

experiments.

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/9184

2.4. Flow cytometry analysis

The degree of fluorescence intensity and the fraction

of labeled SCs were analyzed weekly for 4 weeks byflow cytometry (Elite, Culter, Hialeah, FL). Samples

were gated on forward scatter (FS) vs. side scatter

(SS) to exclude debris and clumps. SCs were analyzed

using a logarithmic amplifier to determine the percen-

tage of stained cells and their mean fluorescence

intensity.

2.5. Phenotypic characteristics of labeled SCs

Indirect immunocytochemical staining was carried

out to assess the phenotypic expression of SC markers

at 1 and 4 weeks after labeling. The following 5 markers

were used: S100 (polyclonal, dilution, 1:100, Dako

Z0331); GFAP (monoclonal, dilution 1:500, SMI 26,

clones Mab1B4, Mab2E1, Mab4A11 and SMI21); P75

(monoclonal, ATCC HB8737, 200-3-G6-4 hybridoma);

major histocompability complex (MHC) I and II(monoclonal, dilution 1:10, Biomeda KO91, clone W6/

32; KO94, clone CR3/43, respectively). For each stain-

ing, SCs were plated on collagen-coated plastic hats,

cultured for 24 h, and washed with PBS. SCs were fixed

with 4% paraformaldehyde and 0.2% Triton-100 for

S100 and GFAP staining but not for P75 and MHC

staining as these are cell surface markers. The SCs were

incubated with antibodies for 30 min followed byincubation with secondary conjugated antibody (Alexa

Fluor 594 goat anti-mouse/rabbit IgG, 1:200, Molecular

Probes, Eugene, OR) for 30 min. Controls SCs (un-

labeled) were incubated with secondary antibody only.

Slides were examined with a confocal laser scanning

microscope 510 (Zeiss, Germany).

2.6. Leakage test

Human SCs labeled with CFSE were co-cultured with

rat SCs (isolated from adult female Fisher rat, 150�/160

g; Charles River Laboratories, Raleigh, NC) on a

collagen-coated hat. After 48 h, a human specific P75

(monoclonal, ATCC HB8737, 200-3-G6-4 hybridoma)

stain was used to distinguish the human SCs from the

non-staining rat SCs. Nuclear staining was carried outat the end of immunostaining with mounting medium

containing Hoechst 33342. Rat SCs will be nuclear

stained by Hoechst, with a negative staining for human

specific P75. CFSE fluorescence observed in any rat SCs

would suggest CFSE leakage from human to rat SCs.

The leakage test was performed in triplicate.

2.7. In vivo human SC transplantation

Human SCs were labeled with 5 mm CFSE and

incubated for 1 min at 37 8C (the optimal labeling

condition established in our in vitro study), then

thoroughly washed and injected into the spinal cord of

adult female Sprague�/Dawley nude rats (Harlan Bio-

product for Science, Indianapolis, IN). The purpose was

to observe the survival, migration of grafted SCs as well

as the stability of the labeling. All surgical procedures

were performed under sterile conditions. Animal care

was provided in accordance with the Laboratory

Animal Welfare Act, Guide for the Care and Use of

Laboratory Animals (NIH), and guidelines issued by the

Animal Care and Use Committee of the University of

Miami. In brief, 9 nude rats with average weights of 150

g were anesthetized with 0.5 ml/kg of rat cocktail (for 1

ml: 43 mg ketamine, 8.6 mg zylazine, and 1.4 mg

acepromazine) and immobilized in a stereotaxic frame.

After a T8 or L1 laminectomy, the dura was incised

longitudinally and the spinal cord was exposed. Two

microliter of the labeled SC suspension (2�/105 cells) in

fresh culture medium was injected at T8 or L1 within the

dorsal columns using a micromanipulator over 30 s. The

cells were injected at a depth of �/1 mm into the spinal

cord using a 5 ml Hamilton syringe and a glass

micropipette/needle. A total of 14 injections at T8 or

L1 (7 injections for each location) were performed in 9

nude rats, including 5 animals received double injections

and 4 received single injections.

2.8. Tissue processing

Animals were sacrificed at different times after

transplantation: three at 1 week (6 injections) and six

at 4 weeks (8 injections). The animals were under deep

anesthesia when perfused with heparinized physiological

saline followed by 4% paraformaldehyde and 0.1%

glutaraldehyde in 0.1 M PBS (pH 7.4). The spinal cords

were removed and post-fixed for 24 h in 4% parafor-

maldehyde at 4 8C. Blocks (15 mm in length) of spinal

cord were cut to include injection sites, which were

marked with charcoal powder at the time of the surgery

for easy identification. Blocks were embedded in poly-

ester wax (Steedman, 1957). Serial 10-mm thick coronal

wax sections were cut on a microtome from each block,

mounted on slides, dried overnight at room tempera-

ture, and stored at 4 8C. After dewaxing and dehydra-

tion, the slides were reviewed with a Zeiss Axiophot

fluorescence microscope.

2.9. Statistical analysis

Data were analyzed statistically using unpaired Stu-

dents t-test and one-way ANOVA with a commercial

software package (GraphPad Instat, San Diego, CA).

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/91 85

3. Results

3.1. Optimal conditions for SC labeling

CFSE-labeled SCs were observed with fluorescence

microscopy to determine the conditions needed for

optimal fluorescence intensity. In quadruplicate testing,

all cell labeling occurred rapidly with intense fluores-cence evenly distributed within the cytoplasm. There

were no differences among the different CFSE concen-

trations and different incubation periods. The experi-

ments demonstrated that with the lowest concentration

(5 mM) used and shortest incubation period (1 min), SCs

were intensely fluorescent labeled compared to the other

concentrations and times tested. Although there were no

significant differences in the cell viability among labeledSCs under different labeling conditions and non-labeled

SCs (P �/0.05), the efficiency of SC plating in culture

decreased with increasing CFSE concentration and

incubation time (Fig. 1). The lower the CFSE concen-

tration used, the higher the plating efficiency. Following

15 min of incubation, plating efficiency decreased

significantly (P �/0.05) at all CFSE concentrations

tested. Thus, the quadruplicate multiple tube testingexperiments demonstrated that an incubation of 1 min

at 37 8C with a concentration of 5 mM of CFSE was the

optimum condition for SCs labeling and this condition

was selected for further in vitro and in vivo testing.

3.2. Flow cytometric analysis of CFSE-labeled SCs

Flow cytometric analysis demonstrated that the

fluorescence intensity of the CFSE, a measure of the

stability of the label, was high during the first week, and

decreased at 2 and 3 weeks, with the signal gradually

disappearing from the SCs at the end of 4 weeks (Fig. 2).

This is consistent with the observations by fluorescence

microscopy of the labeled SCs during the extendedculture period.

3.3. Biological characteristics of the labeled SCs

The cell growth of labeled SCs was relatively slower

than unlabeled SCs during the first week after labeling,

and then it caught up with control cells at the second

week, which was estimated by the days needed to

achieve cell confluence. The purity of the labeled cell

cultures remained at 92�/95% throughout the study

period. Uniformity in shape and size of labeled SCswas similar to unlabeled cultures. S100 staining re-

mained strongly positive for all labeled cells throughout

the 4-week period. There were no obvious differences

microscopically in immunostaining intensity with mar-

kers of GFAP, P75, MHC I and MHC II between CFSE

labeled and non-labeled SCs.

3.4. Leakage testing experiment

The results of triplicate in vitro human and rat SC co-culture experiments demonstrated that there was no

leakage of dye from labeled human SCs to rat SCs (Fig.

3A). The in vivo study also showed that for up to 4

weeks there was no leakage from grafted SCs into the

surrounding cells. This was particularly clear when cells

had migrated into the central canal, where there was no

dye leakage into the adjacent ependymal cells (Fig. 3B).

3.5. Identification of grafted labeled human SCs within

the nude rat spinal cord

Histological analysis demonstrated that the grafted

SCs migrated proximally or distally within the white and

gray matter away from the injection site for a distance

ranging from 3 to 6 mm. Interestingly, in 3 injections the

grafted SCs migrated proximally into the central canal

for up to 12 mm away from the injection site (Fig. 4).

Fluorescence microscope observation demonstrated

that grafted SCs in 12 of the transplants survived andmigrated in the spinal cord of nude rats at 1 and 4

weeks. In two of the spinal cords (T8, L1) from two

animals the grafted SCs could not be recognized at 4

weeks. In the 12 injections, clumped grafted SCs with

fluorescent signal were easily recognized and localized

within the injection site as well as through its migration

path (Fig. 5A and B). At 4 weeks in 6 injections, the

fluorescence intensity decreased and the fluorescencedistribution became uneven. Thus, grafted SCs looked

atrophic, but the Hoechst 33342 and H&E staining

provided evidence that the grafted SC nuclei were still

Fig. 1. Plating efficiency of human SCs following CFSE labeling. The

plating efficiency of SCs decreased with increasing incubation time in

all concentrations of CFSE tested. Following 10 min of incubation the

plating efficiency decreased significantly at all dye concentrations

except the 5 mM. There was also significant decreased plating efficiency

at concentrations 33 and 50 mM compared to 5, 10 and 20 mM

following 1 min incubation (one-way ANOVA, P �/0.05). Each point

in the graph represents the means of percentage of plated SCs. The

error bars represent the standard error of the mean (S.E.M.).

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/9186

intact and that the cells had maintained their normal

structure (Fig. 6).The spinal cord surrounding the grafted SCs were, in

generally, histologically normal. In a few cases, at the 4

weeks point, some tiny vacuoles were observed in the

spinal cord at the injection site. Overall, the animals

transplanted with labeled SCs showed microscopically

no clear evidence of severe inflammation or necrosis

within their spinal cords.

4. Discussion

Transplantation of SCs is a promising treatment

modality to improve neuronal regeneration and axonal

remyelination. Therapeutic use of SC transplantationrequires further studies to evaluate the fate and the

contribution of the transplanted SCs to the axonal

regeneration and remyelination. A major problem with

this therapeutic strategy is the difficulty in identifying

and localizing the fate of engrafted SCs within the

recipient’s nervous system. Therefore, a reliable method

of identifying transplanted SCs is essential for SCs

transplantation studies. Genetic labeling is an advancedmethod for cell identification. Feltri et al. (1992) and

Mosahebi et al. (2000) reported introduction of lacZ

gene in SCs in vitro and in vivo studies. However,

establishing a simpler, reliable and more direct method

to label SCs is still a significant practical task.CFSE, a membrane-permanent dye that covalently

attaches to free amines of cytoplasmic proteins has been

reported to be effective in labeling hepatocytes, lym-

phocytes, CD34� cells, neural stem cell, fetal CNS cells,

and human intervertebral disc cells (Paramore et al.,

1992; Fujioka et al., 1994; Ostrowska et al., 1999;

Hasbold et al., 1999; Gruber et al., 2000; Weijer et al.,

2002; Groszer et al., 2001). Our results demonstrated

that human SCs could also be labeled with 5 mM CFSE

easily and efficiently by incubation at 37 8C for 1 min.

The fluorescence signal diffused evenly throughout the

cytoplasm of SCs and was stable for at least 4 weeks

without leakage. More importantly, the dye was not

toxic to SCs at concentration selected, the labeled SCs

maintained their typical spindle shape and uniform

nucleus with high viability and plating efficiency. The

fluorescence intensity of immunostaining remained high

indicating the unchanged phenotypic expression of these

markers. In vivo studies showed that the grafted SCs

could survive in the nude rat spinal cord for up to 4

weeks. The cells migrated impressive distances from the

implantation site within the white/gray matter or within

the central canal. Thus, SC labeling with CFSE appears

to be relatively easy, nontoxic to cells, and simpler than

genetic labeling methods.

Fig. 2. Histograms representing flow cytometry analysis of human SCs labeled with CFSE at different time points after incubation. The fluorescence

intensity gradually decreased during the period of 4 weeks after labeling. It was almost the same fluorescence intensity between labeled and non-

labeled SCs at the end of the observation period (4 weeks).

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/91 87

We observed that labeled SCs could retain fluores-

cence for at least 4 weeks in culture. We hypothesize the

reason for this limitation, was the continuing dilution of

the fluorescent dye intensity as cells underwent cell

division. As mentioned earlier, CFSE is a fluorescent

dye that penetrates cell membranes and is metabolized

and trapped within the cell. The dye is equally parti-

tioned between daughter cells, so fluorescence intensity

decreases by half with each cell division. Based on this

special property, the technique for analyzing cell divi-

Fig. 3. (A) Leakage test performed with a co-culture of CFSE labeled human SCs and SCs isolated from Fisher rat. After 24 h in culture, human SCs

were double labeled with a human specific P75 antibody to the low affinity nerve growth factor receptor (red) and CFSE labeling (green) plus nuclear

staining with Hoechst 33342. Rat SCs were recognized with their blue nuclear staining by Hoechst 33342, and did not take up any CFSE fluorescence

in their cytoplasm (arrows). (B) Numerous labeled human SCs migrated into the central canal at 1 week after grafting at the T8 level. The host

ependymal cells (arrows), which line the central canal did not pick up any CFSE fluorescence, again suggesting that there was no CFSE leakage from

labeled SCs. Scale bars: (A) 50 mm; (B) 200 mm.

Fig. 4. The grafted human SCs migrated into the central canal of spinal cord in a nude rat (the same rat in Fig. 3B) at 1 week after injection (HE

staining). Single arrow shows the injection site, double and triple arrows show the SCs within the central canal which had migrated a significant

distance (12 mm) from the site of injection. Scale bars, 400 mm.

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/9188

Fig. 5. (A) Grafted human SCs labeled with CFSE survived and migrated along the white and gray matter of spinal cord in a nude rat at 1 week after

transplantation. The CFSE fluorescence in the migrated SCs was intense and the SCs migrated in an organized manner with its normal structure. (B)

The same migrated SCs shown by their nuclear staining with Hoechst 33342. Scale bars, 150 mm.

Fig. 6. Grafted human SCs labeled with CFSE migrated into gray matter of spinal cord in a nude rat after 4 weeks injection. Photographs in A and C

show the uneven distribution of fluorescence in the migrated human SCs labeled with CFSE. H&E staining (B) and nuclear staining with Hoechst

33342 (D) provide evidence that the grafted SCs still had an intact nucleus. Scale bars, 50 mm.

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/91 89

sion using serial halving dilution of the fluorescence

intensity of the vital dye CFSE has become widely used

in immunological laboratories (Lyons, 1999, 2000;

Hasbold et al., 1999). The remarkable fidelity ofpartitioning of this dye allows the clear resolution for

up to 10 sequential division cycles. This approach has

proved suitable for in vitro and in vivo study of B cells,

T cells, NK cells, hemopoietic precursors, as well as cell

lines (Lyons, 2000). In this study, at the 4-week time

point, the grafted SCs, which were not undergoing cell

division, still expressed strong fluorescence, but the

fluorescence did decline slowly over time. In addition,the fluorescence distribution in the cytoplasm was

changed to an uneven pattern, so that the labeled cells

looked atrophic or degenerated when viewed under

fluorescence microscopy. Nevertheless, the grafted SCs

appeared morphologically normal on hematoxylin eosin

staining and Hoechst 33342 nuclear staining. The fact

that there was no leakage of the labeled SCs mixed with

rat SCs in culture established that CFSE, after bindingto intracellular amines, would not label other cells. The

best example was seen in the cells that had migrated into

the central canal and did not stain the host ependymal

cells. This was further demonstrated in our in vivo

study, in which the engrafted SCs had remarkably

distinguishable boundaries and showed no leakage to

surrounding cells. In the two injections where grafted

SCs were missing at 4 weeks, an injection failure duringthe transplantation procedure is the most likely expla-

nation.

As many reports pointed out, the fluorescence in-

tensity measured by flow cytometry is directly propor-

tional to the CFSE concentration during the staining

process, as well as the duration of incubation (Fujioka et

al., 1994; Hasbold et al., 1999; Lyons, 1999). In our

study, by using fluorescence microscopic observation,we concentrated on establishing a labeling condition

with the shortest incubation time to avoid any influence

on the cell characteristics and/or viability for the 4-week

experiment. For a longer study it may require more

intense cell staining to abrogate the fluorescence decay

due to catabolic process or cell division. We believe that

CFSE labeling is also suitable for other neural cells such

as olfactory ensheathing glia, which shares both SC andastrocytic characteristics and is a new candidate for

transplantation to foster repair after spinal cord injury

(Plant et al., 2001). However, initial optimization of

labeling condition with specific cell types and the study

design should be carried out for the best labeling results.

Acknowledgements

We thank Anna Gomez and Yelena Presman for

providing rat SCs, Linda White for her advise on

immunostaining, and Beata Frydel for her assistance

in the confocal laser-scanning microscopic observation

and microphotograph preparation. This study was

supported by The Health Foundation of South Florida.

References

Ansselin AD, Fink T, Davey DF. Peripheral nerve regeneration

through nerve guides seeded with adult Schwann cells. Neuro-

pathol Appl Neurobiol 1997;23:387�/98.

Berry M, Rees L, Hall S, Yiu P, Sivers J. Optic axons regenerate into

sciatic isografts only in the presence of Schwann cells. Brain Res

Bull 1988;20:223�/31.

Butcher EC, Scollay RG, Wetissman IL. Direct fluorescent labeling of

cells with fluorescein or rhodamine isothiocyanate. II. Potential

application to studies of lymphocyte migration and maturation. J

Immunol Methods 1980;37:109�/21.

Casella GT, Bunge RP, Wood PM. Improved method for harvesting

human Schwann cells from mature peripheral nerve and expansion

in vitro . Glia 1996:327�/38.

Evercooren ABV, Gransmuller A, Clerin E, Gumpel M. Hoechst

33342 a suitable fluorescent marker for Schwann cells after

transplantation in mouse spinal cord. Neurosci Lett

1991;131:241�/4.

Feltri ML, Scherer SS, Wrabetz L, Kamholz J, Shy ME. Mitogen-

expanded Schwann cells retain the capacity to myelinate regener-

ating axons after transplantation into rat sciatic nerve. Proc Natl

Acad Sci USA 1992;89:8827�/31.

Felts PA, Smith KJ. Conduction properties of central nerve fibers

remyelinated by Schwann cells. Brain Res 1992;574:178�/92.

Fujioka H, Hunt PJ, Rozga J, Wu GD, Cramer V, Demetriou AA, et

al. Carboxyfluorescein (CFSE) labeling of hepatocytes for short-

term localization following intraportal transplantation. Cell Trans-

plant 1994;3:397�/408.

Groszer M, Erickson R, Scripture-Adams DD, Lesche R, Trumpp A,

Zack JA, et al. Negative regulation of neural stem/progenitor cell

proliferation by the Pten tumor suppressor gene in vivo . Science

2001;194:2186�/9.

Gruber HE, Leslie KP, Ingram JA, Hanley EN, Jr. Optimization of 5-

(and 6-)carboxyfluorescein diacetate succinimidyl ester for labeling

human intervertebral disc cells in vitro . Biotech Histochem

2000;75:118�/23.

Hasbold J, Gett AV, Rush JS, Deenick A, Avery D, Jun H, Hodgkin

PD. Quantitative analysis of lymphocyte differentiation and

proliferation in vitro using carboxyfluorescein diacetate succinimi-

dyl ester. Immunol Cell Biol 1999;77:516�/22.

Hermanns S, Wunderlich G, Rosenbaum C, Hanemann CO, Muller

HW, Stichel CC. Lack of immune responses to immediate or

delayed implanted allogeneic and xenogeneic Schwann cell suspen-

sion. Glia 1997;21:299�/314.

Honig MG, Hume RI. Fluorescent carbocyanine dyes allow living

neurons of identified origin to be studied in long-term cultures. J

Cell Biol 1986;103:171�/87.

Horan PK, Slezak SE. Stable cell membrane labeling. Nature

1989;340:167�/8.

Horan PK, Melnicoff MJ, Jensen BD, Slezak SE. Fluorescent cell

labeling of lymphocytes for studies of cell tracking. Methods Cell

Biol 1990;33:469�/90.

Karrer FM, Reitz BL, Hao L, Lafferty KJ. Fluorescein labeling of

murine hepatocytes for identification after intrahepatic transplan-

tation. Transplant Proc 1992;24:2820�/1.

Levi ADO, Bunge RP. Studies of myelin formation after transplanta-

tion of human Schwann cells into the severe combined immuno-

deficient mouse. Exp Neurol 1994;130:41�/52.

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/9190

Levi ADO. Charaterization of the technique involved in isolating

Schwann cells from adult human peripheral nerve. J Neurosci

Methods 1996;68:21�/6.

Lyons AB. Divided we stand: Tracking cell proliferation with

carboxyfluorescein diacetate succinimidyl ester. Immunol Cell

Biol 1999;77(6):509�/15.

Lyons AB. Analysing cell division in vivo and in vitro using flow

cytometric measurement of CFSE dye dilution. J Immunol

Methods 2000;243:147�/54.

Mosahebi A, Woodward B, Green C, Martin R, Terenghi G. Long-

term effect of vital labeling on mixed Schwann cell cultures.

Histochem J 2000;32:337�/43.

Mosahebi A, Woodward B, Wiberg M, Martin R, Terenghi G.

Retroviral labeling of Schwann cells: in vitro characterization and

in vivo transplantation to improve peripheral nerve regeneration.

Glia 2001;34:8�/17.

Olszewski WL. In vivo migration of immune cells. Boca Raton,

Florida: CRC Press, 1987.

Ostrowska A, Karrer KM, Bilir BM. Histological identification of

purified and cryopreserved allogeneic hepatocytes following trans-

plantation in a murine model without host immunosuppression.

Transplant Int 1999;12:188�/94.

Paramore CG, Turner DA, Madison RD. Fluorescent labeling of

dissociated fetal cells for tissue culture. J Neurosci Methods

1992;44:7�/17.

Plant GW, Ramon-Cueto A, Bunge MB. Transplantation of Schwann

cells and ensheathing glia to improve regeneration in adult spinal

cord. In: Ingoglia NA, Murray M, editors. Axonal regeneration in

the central nervous system. New York: Marcel Dekker, 2001:529�/

61.

Pleasure D, Kreider B, Sobue G, Ross AH, Koprowski H, Sounedfeld

KH, et al. Schwann-like cells cultured from human dermal

neurofibromas. Ann NY Acad Sci 1986;486:227�/40.

Steedman HF. Polyester wax: a new ribboning embedding medium for

histology. Nature 1957;179:1345.

Weijer K, Uittenbogaart CH, Voordouw A, Couwenberg F, Seppen J,

Bianca B, et al. Intrathymic and extrathymic development of

human plasmacytoid dendritic cell precursors in vivo . Blood

2002;99:2752�/9.

Weston SA, Parish CR. New fluorescent dyes for lymphocyte

migration studies. Analysis by flow cytometry and fluorescence

microscopy. J Immunol Methods 1990;133:87�/96.

Xu XM, Guenard V, Kleitman N, Aebischer P, Bunge MB. A

combination of BDNF and NT-3 promotes supraspinal axonal

regeneration into Schwann cell grafts in adult rat thoracic spinal

cord. Exp Neurol 1995a;134:161�/72.

Xu XM, Guenard V, Kleitman N, Bunge MB. Axonal regeneration

into Schwann cell-seeded guidance channels grafted into transected

adult rat spinal cord. J Comp Neurol 1995b;351:145�/60.

Xu XM, Chen A, Guenard V, Kleitman N, Bunge MB. Bridging

Schwann cell transplants promotes axonal regeneration from both

the rostral and caudal stumps of transected adult rat spinal cord. J

Neurocytol 1997;26:1�/16.

X. Li et al. / Journal of Neuroscience Methods 125 (2003) 83�/91 91