Brought to you by Portland Press Ltd.
Published on behalf of the International Federation for Cell Biology
Cancer Cell death Cell cycle Cytoskeleton Exo/endocytosis Differentiation Division Organelles Signalling Stem cells Trafficking
Cell Biology International (2010) 34, 911–915 (Printed in Great Britain)
Flow cytometric sorting and analysis of human epidermal stem cell candidates
Michał Pikuła*1, Karolina Kondej†, Janusz Jaśkiewicz‡, Jarosław Skokowski‡ and Piotr Trzonkowski*
*Department of Clinical Immunology and Transplantology, Medical University of Gdansk, Debinki 1, 80211 Gdansk, Poland, †Department of Plastic Surgery, Medical University of Gdansk, Debinki 7, 80211 Gdansk, Poland, and ‡Department of Surgical Oncology, Medical University of Gdansk, Debinki 7, 80211 Gdansk, Poland

ESC (epidermal stem cells) play a central role in the regeneration of human epidermis. These cells are also responsible for wound healing and neoplasm formation. Efficient isolation of ESC allows their use in medicine and pharmacy as well as in basic science. Cultured keratinocytes and ESC may be used as biological dressing in burn injuries, chronic wounds and hereditary disorders. Therefore, the isolation and characterization of ESC have been goals in biomedical science. Here, we present a flow cytometric method for the isolation and analysis of human ESC candidates. The strategy presented for the isolation of ESC combines previously proposed enzymatic digestion and FACS-sorting of the obtained cell suspension that utilizes morphological features, integrin-β1 expression and Rh123 (Rhodamine 123) accumulation of the cells. We also performed a flow cytometric analysis of sorted cells using a cell tracer.

Key words: cell sorting, cell tracking, epidermal stem cell, keratinocyte

Abbreviations: CFSE, carboxyfluorescein succinimidyl ester, ESC, epidermal stem cells, MDR, multidrug resistance pump, PNA, peptide nucleic acid, Rh123, Rhodamine 123, TAC transit-amplifying cell,

1To whom correspondence should be addressed (email

1. Introduction

The epidermis is a self-renewing tissue, which regenerates continuously. It consists mainly of keratinocytes of varying degrees of differentiation, from the proliferative basal layer to the terminally differentiated horny layer. Looking into cell division kinetics, three subpopulations of keratinocytes have been defined by cell kinetic analysis, that is, stem cells, TACs (transit-amplifying cells) and postmitotic differentiated cells. ESC (epidermal stem cells) represent a minor subpopulation of relatively quiescent slow-cycling cells, defined by their great proliferative potential and unlimited capacity for self-renewal (Barthel and Aberdam, 2005). ESC also play an important role in cellular regeneration, wound healing and neoplasm formation (Morris et al., 2000; Fuchs et al., 2004). ESC constitute between 1% and 10% of the basal layer of epidermis, depending on the methodology used (Morris and Potten, 1994; Heenen and Galand, 1997). It is believed that ESC divide asymmetrically, giving rise to another stem cell and TAC (Koster and Roop, 2005). TACs – the progeny of the stem cells with a limited proliferative capacity can be identified as a pool of rapidly proliferating cells that are lost from the basal layer to terminal differentiation (Jones and Watt, 1993). Cultured human keratinocytes and ESC may be used in autografts as a biological dressing in burn injuries, chronic wounds, soft tissue trauma and various skin diseases. Additionally, ESC have become a target for gene therapy and drug testing (Bannasch et al., 2003; Watt et al., 2006). However, isolation and identification of putative human ESC is still a challenge for today's cell biology and medicine, mainly because of the absence of well-defined cell surface markers of ESC.

One of the best studied markers of ESC are integrins – heterodimeric glycoproteins that are responsible for attachment of basal cells to the basement membrane. The expression of integrin-β1 has been utilized to distinguish between ESC and other keratinocytes (Li et al., 1998; Rossum et al., 2001). However, the majority of the cells of the basal layer in the human epidermis exhibit expression of integrin-β1, and therefore, a new method of ESC isolation should be developed.

Trials with metabolic dyes such as Rh123 (Rhodamine 123) to distinguish ESC from TAC and postmitotic keratinocytes can be helpful. Rh123 labels mitochondria with increasing intensity proportional to cellular activation. The intensity of fluorescence also correlates with the activity of MDR (multidrug resistance pump), for which Rh123 is a substrate. Cells with high expression of MDR are characterized by low fluorescence of Rh123, due to the pumping out of this dye outside the cell (Yeheskely-Hayon et al., 2009). It is believed that ESC as quiescent cells with high expression of the MDR pump do not accumulate Rh123 (Rh123 dim cells) (Drukala et al., 2003). Hence, low accumulation of Rh123 may be utilized as another marker of ESC. Indeed, Rh123 has already been used for the identification of ESC and also human haematopoietic stem cells (Ratajczak et al., 1998; Drukala et al., 2003). ESC are also characterized by small size and low granularity, and therefore, they can be distinguished using the light parameters of a flow cytometer (Zhou et al., 2004).

In the current study, we describe a strategy for ESC isolation and analysis combining an enzymatic step, FACS sorting and a cell tracking method.

2. Materials and methods

2.1. Isolation and culture of human keratinocytes

Aseptic samples of the normal skin from nine surgical patients (the mean age was 47±13.4) were transferred to a plastic vessel with sterile PBS solution containing 500 units/ml penicillin and 0.4 mg/ml streptomycin. The material was stored at 4°C for a maximum of 3 h before the experiment. Immediately before experiments, the samples were washed twice with PBS solution containing penicillin (500 units/ml) and streptomycin (0.4 mg/ml), after which, the subcutaneous fatty tissue was cut off with a sharp blade. The resulting sheets were cut into small fragments, washed with PBS, transferred to 0.5% dispase (Gibco) solution and incubated at 4°C for 18–24 h. Forceps were used to separate the epidermis from dermis and incubated in PBS with 0.25% trypsin (Sigma) 37°C for 20 min. After an additional trypsinization at 37°C for 10 min, trypsin was inhibited by 10% bovine serum, keratinocytes were resuspended by pipetting and harvested by centrifugation at 1500 rev./min for 5 min. The cell suspension obtained was used in further experiments (Fernandez et al., 1998).

Cells directly after FACS sorting were seeded on plastic six-well dishes and cultured for 24 h in KBM (keratinocyte basal medium)/KGM (keratinocyte growth medium) supplemented with 10% FCS (fetal calf serum) in a humidified atmosphere with 5% CO2 at 37°C. The next days, cells were cultured in basal medium without serum (KBM, Clonetics), supplemented with KGM (Clonetics): epidermal growth factor (10 ng/ml), bovine pituitary extract (70 mg/ml), hydrocortisone (0.5 mg/ml), insulin (5 mg/ml), gentamycin (100 mg/ml) and fungizone (0.25 mg/ml). The medium was replaced every other day.

The procedure was approved by the Ethics Committee of the Medical University of Gdansk, and the samples of the skin were only taken when informed consent from the patients was received.

2.2. Staining and cell sorting

Cell suspension obtained during enzymatic digestion was double stained with anti-integrin-β1 antibody and Rh123. APC (allophycocyanin)-conjugated anti-integrin-β1 antibody (IgG1, clone MAR4, Becton Dickinson) (20 μl per 1×106 cells) and Rh123 (Molecular Probes) (final concentration 5.0 μM) were added simultaneously to the cells, and the mixture was incubated for 30 min. Cells were washed with PBS/1% NHCS (normal heated calf serum) (Invitrogen) and centrifuged for 5 min at 1500 rev./min. A FACS Aria cytometer (BD) was used to analyse and sort keratinocyte subpopulations. Light scatter and fluorescence parameters were used to establish sorting gates for epithelial stem cell candidates (ESC) and TAC as described in detail in the Results section.

2.3. Cell kinetics and integrin-β1 expression analysis

Keratinocytes directly after FACS sorting were incubated at 37°C for 15 min in 1 ml of PBS containing CFSE (carboxyfluorescein succinimidyl ester) (Molecular Probes) at 5 μM, a concentration which was determined in presumptive experiments. After two washing steps in PBS containing 1% heat-inactivated bovine serum, cells were plated at a density of 1×103/cm2 and cultured as described above. After 14 days of incubation in six-well plates (BD) at 37°C and 5% CO2, CFSE-labelled cells were washed twice in PBS and stained with anti-integrin-β1 antibody (IgG1, clone MAR4, BDBioscience). Stained samples were analysed using a LSRII (BD) cytometer and WinMdi software (kindly provided by Dr J. Trotter).

2.4. Flow cytometric analysis of involucrin and integrin-β1 expression

Cell suspension obtained during enzymatic digestion was double stained with anti-integrin-β1 and anti-involucrin antibodies. Freshly isolated cells were washed in PBS, centrifuged and the pellet (2×105 cells/sample) was suspended in the permeabilization buffer (0.25% saponin in PBS with 0.1% BSA) and the primary mouse anti-human involucrin antibody (1:50) (IgG1, clone SY5, Novocastra). Unstained cells and cells stained with isotype control antibody were used as controls. After 3 h of incubation, the cells were washed twice and resuspended in the permeabilization buffer with secondary sheep anti-mouse FITC-conjugated polyclonal antibody (NCL-SAM-FITC, Novocastra) (1:50). After 1 h, the cells were washed and resuspended again in the permeabilization buffer containing APC-conjugated anti-integrin-β1 antibody (IgG1, clone MAR4, Becton Dickinson). After 1 h of incubation, the cells were washed, resuspended in PBS and analysed in an LSRII flow cytometer (Becton Dickinson). There were 10000 cells collected from each sample. Fluorescence signal intensities were analysed with FACSDiva Software (BD Biosciences), and graphical representations of the signal were prepared with WinMdi 2.8 (software kindly provided by Dr J. Trotter).

2.5. Telomere length measurement by Flow-FISH (fluorescence in situ hybridization)

Telomeric sequences in keratinocytes were detected with Flow-FISH telomere kit (DAKO, K5327). Keratinocytes were washed in PBS, centrifuged and the resulting pellet (5×105 cells/sample) was suspended in 0.1% BSA in PBS with APC-conjugated anti-integrin-β1 antibody (Becton Dickinson). After 1 h of incubation, keratinocytes were washed, mixed with 1301 cells (ATCC) and incubated in a hybridization solution containing FITC-labelled telomere-specific PNA (peptide nucleic acid) probe for 10 min in a thermoblock (82°C). As a control, the hybridization solution without PNA probe was used. After overnight incubation in the dark at room temperature, cells were heated to 40°C for 10 min, washed and resuspended in PBS containing 10 μg/ml of propidium iodide and 100 μg/ml RNAse. Samples were incubated for another 2 h at room temperature in the dark and analysed on the flow cytometer (LSR II, BD). Telomere fluorescence was defined as the mean of FITC fluorescence of the G0/G1 keratinocytes with high and low expression of integrin-β1. The relative length of telomeres was calculated as the ratio between mean sample telomere fluorescence and mean telomere fluorescence measured in 1301 cells, with correction for the DNA index of the G0/G1 cells multiplied by 100. Signal intensities were analysed with FACSDiva Software (BD Biosciences, USA).

2.6 Statistics

Data were computed using the software Statistica 7.0 (Statsoft). The analysis of data obtained from the patients was based on parametric tests as indicated by data distribution. P<0.05 was recognized as significant.

3. Results

3.1. High expression of integrin-β1 is a marker of ESC candidates

Co-staining of isolated epithelial cells for integrin-β1 with involucrin revealed that the subset with high expression of integrin-β1 was negative for involucrin (Figure 1A). In the integrin-β1 bright population, there were 1.26%±0.50 of involucrin-positive cells in comparison with 6.26%±1.28 of involucrin-positive cells in the integrin-β1 dim keratinocytes population (t test: P = 0.019). The analysis of telomere length of isolated keratinocytes revealed that integrin-β1 bright cells were characterized by significantly longer telomeres than those integrin-β1 negative (Figure 1B) (t test: P = 0.034). Both low expression of involucrin and preserved telomeres found in the integrin-β1 bright population are characteristic features of undifferentiated cells, which fits the definition of quiescent ESC candidates. Hence, integrin-β1 expression could be used as a marker in the FACS-sorting procedure.

3.2. Staining and sorting of the isolated epithelial cells distinguish between two subsets with expression of integrin-β1

Immediately after enzymatic isolation, a mixture of skin cells was labelled with Rh123 and integrin-β1 antibodies. There were a few steps in the FACS sorting procedure. First, a gate was established on the basis of the size and granularity of the cell suspension obtained during enzymatic isolation using the FSC (forward scatter) compared with the SSC (side scatter) dotplot (gate P1 in Figure 2A). The gate reduced heterogeneity of the population obtained during the enzymatic step of the procedure as it covered only the cells from the region with relatively low values of the FSC and SSC parameters. The gate contained around 50% of the total population. This approach allowed the exclusion from further sorting of acellular debris represented by the events with the lowest FSC and SSC (mainly cell membranes from dead cells), as well as cell conglomerates represented by the events with the highest FSC and SSC (resulting mainly from uncompleted enzymatic digestion of epithelium and contamination with other cell types from the skin in the conglomerates). Still, the gated population included mixture of ESC candidates and TAC.

In the next step, the discrimination between ESC and TAC was performed. Staining of this mixed population with Rh123 revealed the presence of two populations of cells with high and low Rh123 accumulation. At the same time, these cells demonstrated a very wide range of expression of integrin-β1, from weak or negative (negative control level) to a very strong signal. Dotplot of Rh123 compared with integrin-β1 from the above described P1 gate allowed the establishment of three regions for FACS sorting (Figure 2B). Gates P2 and P3 represented integrin-β1 bright cells with low or high accumulation of Rh123, respectively. Integrin-β1 bright cells with low accumulation of Rh123 are classically regarded as a population enriched with candidates of ESC. Integrin-β1 bright cells with high accumulation of Rh123 seem to have a higher rate of metabolism than those with low Rh123 accumulation, which supposedly made them less accurate as ESC candidates. The P4 gate represented cells with medium expression of integrin-β1 and high accumulation of Rh123 that were TAC candidates.

3.3. Sorted epithelial cells reveal two different patterns of growth in vitro

Analysis of the distribution of the cell tracer (CFSE) demonstrated that the kinetics of cell division is different between ESC and TAC candidate subsets (Figure 2C). Compared with TAC cells (from P4 gate), ESC candidates (from P2 and P3 gates) exhibited a slower rate of cell division noted as a more intense peak of CFSE fluorescence. The fluorescence signal of CFSE was almost identical between gates P2 and P3. In addition, the cells from P2 and P3 gates contained a fraction which did not divide in the culture at all (the fluorescence signal after 14 days was similar to the positive control). Such a non-dividing fraction could not be found in the culture of TAC cells. All TAC cells from the P4 gate divided after 14 days of cell culture.

4. Discussion

In the current study, we found that enzymatic digestion combined with multiparameter FACS sorting allows distinguishing and sorting of ESC candidates, which can then be cultured in vitro. High expression of integrin-β1 and the characteristic morphology of these cells were helpful in defining proper sorting parameters for ESC. The identification was then confirmed by revealing slow cyclic kinetics of sorted cells in a CFSE assay, which is a characteristic feature of ESC.

Numerous data suggest that the longest telomeres are a general feature of different adult stem cell compartments (Flores et al., 2008). Analysis of the cultured keratinocyte samples selected in this study showed heterogeneity in the relative telomere length. Cells with high expression of integrin-β1 revealed significantly longer telomeres than other keratinocytes. Another feature of ESC is their undifferentiated state and the absence of some proteins characteristic of the suprabasal layer of epidermis, for example cytokeratins 1, 10 and involucrin (Bikle et al., 2001). Involucrin is regarded as a marker of keratinocyte differentiation. The protein is a substrate for the keratinocyte transglutaminase and a component of the cornified envelope. Prior to envelope assembly, it is present as a soluble cytoplasmic protein and can be detected by immunocytochemistry (Simon and Green, 1988). Hence, the absence of involucrin together with longer telomeres in the integrin-β1 bright cells confirmed that high expression of integrin-β1 can be used as a marker of ESC.

Here, we tried to develop a procedure combining morphological criteria with integrin-β1 expression and Rh123 accumulation analysis. Studies of other authors demonstrated that ESC are characterized by low values of FSC and SSC that correspond to the low level of differentiation of keratinocytes. The process of differentiation results in larger keratinocytes with more granules in their cytoplasm. Therefore, in our experiments, initially, discrimination of ESC and TACs relied on gating of the cells with small size and low granularity.

It is known that isolation of cells by proteolytic enzyme treatment can influence antigen detection. This phenomenon mainly concerns surface antigens (Bauer et al., 1994). Nevertheless, the level of integrin-β1 detected after enzymatic digestion in our study was not an obstacle, and the expression was preserved enough to sort separately ESC and TAC from the skin. Surprisingly, we found that the cells with the highest expression of integrin-β1 kept the characteristics of ESC candidates regardless of the Rh123 accumulation. At the same time, Rh123 bright cells with low expression of integrin-β1 possessed characteristics of TAC cells. The distinction between ESC and TAC in the peak of cells with high accumulation of Rh123 allowed us to increase the yield of stem cells in our sorts. Importantly, we did not lose the population with very high expression of integrin-β1 and high accumulation of Rh123. This is of special interest for practical purposes because the initial number of ESC determines the final yield of cells from culture in vitro (Jones et al., 1995). Obviously, the clinical efficacy also depends on the cultivation and transplantation of ESC (Pellegrini et al., 1999).

In this study, we used a CFSE dye to analyse kinetics of cell division. This dye is mainly used to track divisions of lymphocytes. Here, we demonstrated that the CFSE assay may be considered as a method for determining the intensity of cell divisions of human keratinocytes. This technique allowed the observation of subtle differences in the kinetics. In our study, ESC exhibited slower kinetics in comparison with TAC cells. Remarkably, we detected the presence of a fraction of cells in ESC candidate subsets, which did not divide at all in cell culture. These cells may constitute primitive quiescent ESC. Obviously further studies, e.g. quantitative assessment of colony forming efficiency, are necessary to support our finding.

In conclusion, our strategy proved to be a valuable approach allowing the sorting of epidermal cells into an actively cycling TAC compartment and a quiescent ESC compartment. Combining flow cytometric sorting, cell culture and cytometric analysis provides powerful tools for phenotypical and functional characterization of epidermal subpopulations.

Author contribution

Michał Pikuła carried out the conception and design, experiments, data analysis and manuscript writing. Karolina Kondej and Jarosław Skokowski contributed to provision of study materials and data analysis. Janusz Jasśkiewicz contributed to the management of the project and provision of study materials. Piotr Trzonkowski participated in data interpretation, manuscript writing and coordination of the study.


This work was supported by the Polish Ministry of Science and Higher Education, grant N N403 089335. This work was also supported by the HOMING programme of the Foundation for Polish Science (a grant from Iceland, Liechtenstein and Norway through the EEA Financial Mechanism to Piotr Trzonkowski).


Bannasch, H, Fohn, M, Unterberg, T, Bach, AD, Weyand, B and Stark, GB (2003) Skin tissue engineering. Clin Plast Surg 30, 573-9
Crossref   Medline   1st Citation  

Barthel, R and Aberdam, D (2005) Epidermal stem cells. J Eur Acad Dermatol Venereol 19, 405-13
Crossref   Medline   1st Citation  

Bauer, KD and Jacobberger, JW (1994) Analysis of intracellular proteins. In Darzynkiewicz Z, Robinson JP, Crissman HA, editors, Flow cytometry pp. 351-76, San Diego, Academic Press
1st Citation  

Bickenbach, JR and Roop, DR (1999) Transduction of a preselected population of human epidermal stem cells: consequences for gene therapy. Proc Assoc Am Physicians 111, 184-89
Crossref   Medline   

Bikle, DD, Ng, D, Tu, CL, Oda, Y and Xie, Z (2001) Calcium- and vitamin D-regulated keratinocyte differentiation. Mol Cell Endocrinol 177, 161-71
Crossref   Medline   1st Citation  

Drukala, J, Majka, M and Ratajczak, M (2003) Advances in methods of isolation and expansion of human epidermal stem cells. Adv Cell Biol 30, 37-48
1st Citation   2nd  

Fernandez, G, Bejar, JM, Alonso-Varona, A, Masdevall, G and Gabilondo, FJ (1998) Study of the human keratinocyte isolation methods and in vitro culture techniques in a single laboratory. Eur J Plast Surg 21, 353-57
Crossref   1st Citation  

Flores, I, Canela, A, Vera, E, Tejera, A, Cotsarelis, G and Blasco, MA (2008) The longest telomeres: a general signature of adult stem cell compartments. Genes Dev 22, (5), 654-67
Medline   1st Citation  

Fuchs, E, Tumbar, T and Guasch, G (2004) Socializing with the neighbors: stem cells and their niche. Cell 116, (6), 769-78
Medline   1st Citation  

Heenen, M and Galand, P (1997) The growth fraction of normal human epidermis. Dermatology 194, 313-17
Crossref   Medline   1st Citation  

Jones, PH and Watt, FM (1993) Separation of human epidermal stem cells from transit amplifying cells on the basis of differences in integrin function and expression. Cell 73, 713-24
Crossref   Medline   1st Citation  

Jones, P, Harper, S and Watt, FM (1995) Stem cell patterning and fate in human epidermis. Cell 80, 83-93
Crossref   Medline   1st Citation  

Koster, MI and Roop, DR (2005) Asymmetric cell division in skin development: a new look at an old observation. Dev Cell 9, 444-46
Crossref   Medline   1st Citation  

Li, A, Simmons, PJ and Kaur, P (1998) Identification and isolation of candidate human keratinocyte stem cells based on cell surface phenotype. Proc Natl Acad Sci USA 95, 3902-7
Crossref   Medline   1st Citation  

Morris, RJ (2000) Keratinocyte stem cells: targets for cutaneous carcinogens. J Clin Invest 106, 3-8
Crossref   Medline   1st Citation  

Morris, RJ and Potten, CS (1994) Slowly cycling (label-retaining) epidermal cells behave like clonogenic stem cells in vitro. Cell Prolif 27, 279-89
Crossref   Medline   1st Citation  

Pellegrini, G, Ranno, R, Stracuzzi, G, Bondanza, S, Guerra, L and Zambruno, G (1999) The control of epidermal stem cells (holoclones) in the treatment of massive full-thickness burns with autologous keratinocytes cultured on fibrin. Transplantation 68, 868-79
Crossref   Medline   1st Citation  

Ratajczak, MZ, Pletcher, ChH, Marlicz, W, Machalinski, B, Moore, J and Wasik, M (1998) CD34+, Kit+, Rhodamine 123(low) phenotype identifies a marrow cell population highly enriched for human hematopoietic stem cells. Leukemia 12, 942-50
Crossref   Medline   1st Citation  

Rossum, MM, Schalkwijk, J, van de Kerkhof, PC and van Erp, PE (2001) Immunofluorescent surface labelling, flow sorting and culturing of putative epidermal stem cells derived from small skin punch biopsies. J Immunol Methods 267, 109-17
Crossref   1st Citation  

Simon, M and Green, H (1988) The glutamine residues reactive in transglutaminase-catalyzed cross-linking of involucrin. J Biol Chem 263, 34
1st Citation  

Watt, FM, Celso, CL and Silva-Vargas, V (2006) Epidermal stem cells: an update. Curr Opin Genet Dev 16, 518-24
Crossref   Medline   1st Citation  

Yeheskely-Hayon, D, Regev, R, Katzir, H and Eytan, GD (2009) Competition between innate multidrug resistance and intracellular binding of rhodamine dyes. FEBS J 276, 637-48
Crossref   Medline   1st Citation  

Zhou, JX, Chen, SY, Liu, WM, Cao, YJ and Duan, EK (2004) Enrichment and identification of human ‘fetal’ epidermal stem cells. Hum Reprod 19, 968-74
Crossref   Medline   1st Citation  

Received 2 April 2010/6 May 2010; accepted 1 June 2010

Published as Cell Biology International Immediate Publication 1 June 2010, doi:10.1042/CBI20100223

© The Author(s) Journal compilation © 2010 Portland Press Limited

ISSN Print: 1065-6995
ISSN Electronic: 1095-8355
Published by Portland Press Limited on behalf of the International Federation for Cell Biology (IFCB)