Optimization of the Reconstruction of Dermal Papilla ... - Springer Link

Biotechnology and Bioprocess Engineering 2010, 15: 182-190 DOI/10.1007/s12257-009-3050-z

=

Optimization of the Reconstruction of Dermal Papilla like Tissues Employing Umbilical Cord Mesenchymal Stem Cells Bo-Young Yoo1, Youn-Ho Shin2, Hee-Hoon Yoon3, Young-Kwon Seo3, Kye-Yong Song2, and Jung-Keug Park1,3,4* 1

Department of Chemical and Biochemical Engineering, Dongguk University, Seoul 100-715, Korea 2 Department of Pathology, Chung-Ang University, Seoul 140-757, Korea 3 Dongguk University Research Institute of Biotechnology (DURIB), Seoul 100-715, Korea 4 Department of Medical Biotechnology, Dongguk University, Seoul 100-715, Korea

Abstract Alopecia is not life threatening, but patients who undergo alopecia often experience severe mental stress. In addition, the number of individuals afflicted by alopecia has been increasing steadily. The most effective treatment of alopecia developed to date is auto hair transplantation. To overcome the limitations associated with current therapies for the treatment of alopecia, many researchers have attempted to revive hair follicles by áå=îáíêç culture of hair follicle cells and subsequent implantation in the treatment area. Previously, we demonstrated that umbilical cord-derived mesenchymal stem cells (UC-MSCs) could be isolated and expanded successfully from the Wharton’s Jelly. Cultureexpanded UC-MSCs formed aggregates similar to native dermal papilla (DP) in special media (DPFM) and reconstructed dermal papilla like tissues (DPLTs) could induce new hair follicles. The purpose of the present study was to optimize the reconstruction of DPLTs. As in the case of MSCs, when compared to differentiated cells, DPLTs require an additional step to induce differentiation into dermal papilla cells. However, it is necessary to use hepatocyte growth factor (HGF) in the differentiation step, which is relatively expensive. To reduce the expenses associated with cell therapy using MSCs, it is necessary to optimize this differentiation step. To accomplish this, we evaluated the effects of cell inoculation density and growth factors during differentiation. © KSBB = = = =

hÉóïçêÇëW=ãÉëÉåÅÜóã~ä=ëíÉã=ÅÉääI=Ü~áê=ÑçääáÅäÉI=ÇÉêã~ä=é~éáää~I=ÅÉää=íÜÉê~éóI=ìãÄáäáÅ~ä=ÅçêÇ=

INTRODUCTION A hair is a specialized outgrowth of the epidermis that can be divided into two distinct parts, the hair follicle and the hair shaft. The hair follicle is a small, curved pit buried deep in the fat of the scalp that is the point from which the hair grows. The hair follicle is divided into two major classes by epidermal tissues and dermal tissues [1]. The epidermal tissues of hair follicles include the outer root sheath (ORS) and the matrix, and it has been reported that epidermal stem cells are present in the ORS layer [2-4]. The dermal tissue of hair follicles that originate from the mesenchyme contain a dermal papilla (DP) and a dermal sheath (DS), which have a cell pattern similar to dermal fibroblasts [5]. The DP is lo*Corresponding author Tel: +82-2-2260-8535 Fax: +82-2-2271-3489 e-mail: [email protected]

cated in the lower area of the hair follicle, where it maintains hair growth of the hair shaft through complex and continuous interactions with hair matrix cells, while the DS surrounds the entire hair follicle. Among them, the DP has been reported to play a basic role in induction of the formation of the hair follicle, maintenance of hair shaft growth, differentiation of undifferentiated cells in the shaft, and control of the hair cyclic activity [6,7]. Development of hair follicles begins towards the end of the first trimester of pregnancy and is controlled by epidermal-me-senchymal interaction (EMI), which is a signaling cascade between epidermal and mesenchymal cell populations. Alopecia is not life threatening, but patients who undergo alopecia often experience severe mental stress. In addition, the number of individuals afflicted by alopecia has been increasing steadily. Although drug or natural substance therapy may retard the progress of alopecia or prevent future hair loss, it may also accelerate hair loss when the treatment is stopped after prolonged use.

Biotechnol. Bioprocess Eng. 183=

The most effective treatment of alopecia that has been developed to date is auto hair transplantation. Conversely, the transplantation of human hair involves removing plugs of natural hair from areas in which occipital hair is growing and transplanting them to bald areas. Although the transplanted hair settles at the transplant area as a complete hair follicle and becomes a permanent hair that undergoes a normal growth cycle, the amount of hair that can be transplanted is severely limited, with the transplantation of only 2,000 hairs per operation and three operations being possible. To overcome this problem, many researchers have attempted to revive hair follicles by propagating hair follicle cells or mesenchymal cells in vitro and then implanting them in the treatment area. In addition to hair follicular cells, various cytokines and growth factors participate in the maintenance of hair follicles, the growth of the hair shaft, and the control of hair cyclic activity [8]. For example, hepatocyte growth factor (HGF) stimulates the growth of hair follicles in vivo and in vitro [916], while basic fibroblast growth factor (bFGF) stimulates the growth of DP cells in vivo [17-21] and vascular endothelial growth factor (VEGF) stimulates the growth of hair follicles and hair shafts in vivo [22-25]. These cytokines and growth factors are released or controlled by hair follicles as well as supplied to hair follicles by the blood. The purpose of this study was to optimize the reconstruction step of dermal papilla like tissues. As with MSCs, DPLTs require an additional differentiation step to produce DP cells for cell therapy. However, to induce differentiation, it is necessary to use hepatocyte growth factor (HGF), which is relatively expensive. To reduce the costs of cell therapy using MSCs, it is necessary to optimize the differentiation step. To accomplish this, we evaluated the effects of cell inoculation density and growth factor during differentiation. Specifically, we first set the cell inoculation density at 1.6 × 104 cells/mL and then evaluated the effects of various growth factors at this density. Next, selected growth factors (HGF, EGF, and NGF) were tested for their suitability for use in the reconstruction of DPLTs.

A

B

C

D

Fig. 1. Characterization of cells obtained from human umbilical cord mesenchymal stem cells. (A) Inverted microscopic image of primary cultured hUC-MSCs (magnification × 100). Flow cytometry analysis of human umbilical cord mesenchymal cell markers (b~d). (B) CD44 as a hematopoietic cell marker. (C) CD90 as a mesenchymal stem cell marker. (D) CD105 as a mesenchymal stem cell marker.

were attached onto 100 mm tissue culture dishes by contacting the connective tissue onto the dish surface. After 2 h of incubation at 37oC, culture medium (DMEM (low glucose)/10% FBS) was added. After 3~5 days of primary culture, hUC-MSCs appeared. After 7 days of culture, the hUCMSCs were too close together; therefore, they were detached with Accutase and then transferred to new culture dishes at a split ratio of one to five. The initial hUC-MSCs seeding density was 4 × 103 cells/cm2 at the 25 T cell culture flasks [26]. During subculture, the hUC-MSCs were passaged every 5 days. Surface antigen expression was detected using FACS against CD45 (leukocyte marker), CD90, and CD105 (some of mesenchymal stem cell makers).

MATERIALS AND METHODS

Reconstruction of Dermal Papilla-like Tissues (DPLTs)

Primary Culture of Human Umbilical Cord Mesenchymal Stem Cells

Human mesenchymal stem cells originating from the umbilical cord were subjected to a monolayer culture in DMEM supplemented with 10% FBS until the cells occupied approximately 80% of the culture dish. Next, the culture medium was replaced with dermal papilla-forming medium (DPFM) that contained 10 ng/mL hydrocortisone, 10 µL/mL × 100 ITS liquid media (Sigma), and 50 µg/mL gentamicin [27]. Additionally, various growth factors were added to the culture medium in the different experimental groups. Specifically, the control group consisted of cells cultured in DPFM without any supplemental growth factor. The treatment group consisted of cells cultured in DPFM supplemented with 20 ng/mL HGF (rhHGF, R&D system, USA), 10 ng/ mL recombinant human epidermal growth factor (EGF,

Human umbilical cords were obtained from consenting patients delivering full-term (38~40 weeks) infants by Cesarean section. Each human umbilical cord was washed 3 times with Ca2+, Mg2+-free Dulbecco’s phosphate balanced solution (D-PBS) and then massaged to remove blood in the vessels. The cord was then transected longitudinally along the umbilical cord vein. Part of the Wharton’s jelly, including the smooth muscle layer, was clipped with the surgical mess. Next, the two arteries that were exposed were removed using forceps. For the complete removal of blood vessels, most connective tissue was removed by clipping with scissors. For explants outgrowth, six pieces (each 2 mm × 2 mm)

184

A

B

C

D

Fig. 2. Change in morphology during the reconstruction DPLTs step. (A) The morphology of proliferating MSCs in low glucose DMEM with 10% FBS. (B) Preconditioning of MSCs in DPFM for 3 days. (C) Self-aggregated DPLTs formed in the DPFM, but could not be suspended after 4 days. (D) DPLTs were suspended in DPFM after 1 week.

Sigma, USA), or 10 ng/mL nerve growth factor (NGF, R&D system, USA). The medium was replaced with fresh medium at about 3-day intervals for 3 weeks, after which the culture was treated with Accutase (Innovative Cell Technologies Inc., San Diego, CA, USA) at concentrations ranging from 20 to 40 µL/cm2 to detach the cells from the culture dish [2]. Cell aggregation was initiated 24 h after treatment with Accutase, and the completely aggregated cell mass became suspended in the medium. The suspended cell aggregates were then isolated from the culture by centrifugation at 500 rpm for 3 min. Next, the cell aggregates were re-suspended in fresh medium and kept in a culture container to preserve their culture state until they were used in the following experiment. Fig. 2 illustrates the procedure for the formation of dermal papilla like tissues (DPLTs) from MSCs originating from the umbilical cord in vitro. RT-PCR

Total RNA from all three cell types was isolated using Trizol reagent (Invitrogen, CA, USA) according to the manufacturer’s instructions. cDNA was then synthesized by reverse transcription (RT) using 1 µg of the total RNA. PCR was conducted by subjecting the samples to 35 cycles (within the linear range of amplification) of denaturation (94.8oC, 1 min), annealing (52.8oC, 1 min), and extension (72.8oC, 1 min). The products were then analyzed by electrophoresis on 2% agarose gels and subsequent visualization by ethidium bromide staining. We tested well known markers (type 4 collagen, laminin, heat shock protein 70, mitochondrial ribosomal protein S7, and versican) of human DPCs. The cell and tissue extracts were prepared using Pro-prep protein extraction solution (Intron, Daejeon, Korea). Protein samples were run on 15% sodium dodecyl sulfate (SDS)polyacrylamide gel, transferred onto nitrocellulose membrane, and then incubated with anti-human NGAL polyclonal antibody. For an internal control, anti- beta actin antibody (Sigma, St Louis, MO, USA) was used. Blots were

then incubated with peroxidase-conjugated secondary antibody and developed by enhanced chemiluminescence (Amersham, Buckinghamshire, UK). Western Blot

The cell and tissue extracts were prepared using RIPA buffer (RIPA buffer (Radio Immuno Precipitation Assay buffer, Thermo Fisher Scientific Inc. Rockford, USA). Protein samples were run on 15% sodium dodecyl sulfate (SDS)polyacrylamide gel, transferred onto nitrocellulose membrane, and incubated with anti-human versican antibody (ab 19345, Abcam, Denmark), anti-laminin 1+2 antibody (ab 7463, Abcam, Denmark), anti human polygonal to Collagen IV (ab 6586, Abcam, Denmark) antibody. For an internal control, anti-beta actin antibody (Sigma, St Louis, MO, USA) was used. Blots were then incubated with peroxidaseconjugated secondary antibody and developed by enhanced chemiluminescence (Amersham, Buckinghamshire, UK). Immunohistochemistry of DPLT

To determine the size and general structure of the DPLTs, suspended DPLTs were recovered with media and then centrifuged at 500 rpm for 3 min. The supernatant was then discarded and the pellets were mixed with 0.5% collagen gel. After gelation, collagen gel including the DPLTs was fixed in 4% paraformaldehyde for 1 h. For immunohistochemical examination, the ABC (avidin-biotin complex) technique was used with diaminobenzene (DAB) as a chromogene. For detecting hair follicle inducing activity, we used versican (Biovendor, MC955, Czech Republic). Statistical Analysis

Data were reported as the means ± standard deviation and were analyzed using a Student’s t-test. The difference between the means was considered significant when p < 0.05.


A

B

C

D

Fig. 3. Effect of initial cell inoculation density on reconstruction of DPLTs during DPLT morphogenesis. (A) 0.8 × 104 cells/cm2, (B) 1.6 × 104 cells/cm2, (C) 2.4 × 104 cells/cm2 and (D) 4 × 104 cells/cm2.

Fig. 4. The number of reconstructed DPLTs formed when different initial cell inoculation densities were used for DPLT morphogenesis. (A) 0.8 × 104 cells/cm2, (B) 1.6 × 104 cells/cm2, (C) 2.4 × 104 cells/cm2, and (D) 4 × 104 cells/ cm2 (é > 0.05).

RESULTS AND DISCUSSION

days, cell migration and some cell aggregation occurred (Fig. 2C). Finally, reconstructed DPLTs were observed after 5 days (Fig. 2D).

Primary culture of human Umbilical Cord Mesenchymal Stem Cells

Optimization of Cell Inoculation Density for Reconstruction of DPLTs

HUC-MSCs were isolated pure without other cells from the Wharton’s jelly extracted from human umbilical cords. The cells showed a spindle shape similar to skin fibroblasts. Flow cytometric analysis of the surface antigen expression in the UC-derived cultures was performed using the labeled anti CD44-PE, anti CD90-FITC, and anti CD105-R-PE. The HUC-MSCs were positive against the general bone marrowderived mesenchymal stem cell markers, CD90 and CD105, but negative against CD44 (Fig. 1).

MSCs were plated into tissue culture flasks in expansion medium at a density of 0.8~4 × 104 cells/cm2. In all experiments, we detected a large amount of DPLTs within 5 days of detach-attach step. This phenomenon indicates that the UC-MSC differentiated to DP like cells, after which they developed aggregative activity. In addition, cell mobility developed during the pre-treatment and post-treatment steps. These changes resulted in the cell-cell adhesion force becoming stronger than the cell-matrix adhesion force, which facilitated reconstruction of the DPLTs. When the inoculation density was 1.6 × 104 cells, DPLTs formed within 4 days, while they were detected at 6 days after the detachattach step when the inoculation density was 2.4 × 104 cells/ cm2 in control medium. When the inoculation density was 0.8 × 104 cells/ cm2, cell migration was detected even though aggregation was not. When the initial density was 4 × 104 cells/cm2, aggregates were formed, but their size was irregular and they were too large due to joining of the aggregates (Fig. 3D). In the case of high cell density, some of DPLTs lump together. Based on these results, the optimal inoculation cell density ranged from 1.6 × 104 to 2.4 × 104 cells/cm2. The number of reconstructed DPLT ranged from 24 to 86 when the cell inoculation density was varied without altering any other conditions (Fig. 4). The concentration of reconstructed DPLTs per 25 T flask increased as the cell inoculation density increased to 2.4 × 104 cells/cm2; however, the size of the DPLTs was larger than those produced at lower cell densities (under 2.4 × 104 cells/cm2).

Reconstruction of the Dermal Papilla-like Tissues (DPLTs)

The goal of this study was to optimize the cell inoculation density and detach-attach step in DPLT reconstruction using various cell densities. Images of the corresponding cell cultures were taken after cell seeding using phase contrast microscopy (Fig. 2). The expansion medium contained Dulbecco’s Modified Eagles Medium (DMEM, low glucose; Invitrogen Co.) and 10% fetal bovine serum (FBS; Cambrex Co.). Upon reaching 80% confluence, the expansion medium was exchanged with Dermal Papilla Forming Medium (DPFM; differentiation medium). When reconstructing DPLTs employing MSCs, an additional step was required when compared to reconstruction of DPLTs using DPC. However, when pretreatment was conducted for 3 weeks, DPLTs formed within 5 days. After the cell detach-attach process in the same culture flask, the cell distribution changed. After an additional 2

186

A

B

A

E

C

D

B

F

C

G

D

H

Fig. 5. Phase contrast microscopic image of reconstructed DPLTs as an addictive growth factor in DPFM after 7days. (A) No additive growth factor in DPFM, (B) 10 ng/mL rhEGF, (C) 20 ng/mL of rhHGF, and (D) 10 ng/mL NGF.

Effect of Growth Factor during the Formation of DPLTs

HGF is a paracrine factor in the dermal papilla that mediates epidermal-mesenchymal interactions and stimulates follicle growth in vitro [28]. HGF is a multifunctional polypeptide [29,30] that acts as a mitogen [31], motogen, or morphogen [29] in a variety of organs. Hepatocyte growth factor (HGF) may be important for hair follicle morphogenesis. HGF has been shown to be expressed in DP cells and to have an anagen inducing effect in vivo and in vitro [10, 11,16]. EGF is another growth factor that has proliferative effects on cells of both mesodermal and ectodermal origin, particularly keratinocytes and fibroblasts. As a promoter of epithelial cell growth, EGF may also play an important role in hair follicle embryogenesis. Furthermore, it has been shown that there is reduced expression of EGF-R in epidermal cells overlying the hair germ [32,33]. Later stages of follicle maturation show EGF and EGF-R expression, but the specific stage for the onset of initial hair follicle expression has not yet been determined [32-34]. Injection of EGF promotes epidermal thickening and inhibition of hair fiber growth [35]. NGF and neutrophin-4/brain derived neutrophic factor (NT-4/BDNF) may play both overlapping and complementary roles in early follicular development. The former appears to promote the proliferation of mesenchymal cells before they become organized into a follicular structure. This mechanism may lead to NGF contributing to the growth of primordial follicles. The mechanisms by which NGF and NT-4/ BDNF promote follicular development are unknown [36,37]. Application of 10 ng/mL rhEGF, 20 ng/ml rhHGF, and 10 ng/mL NGF in DPFM resulted in self-aggregated DPLTs. During reconstruction of DPLTs using HUC-MSCs, we compared the effect of added growth factor on the DPLT at a constant initial cell inoculation density. When compared to the control group (without any growth factor), the recon-

Fig. 6. Histochemical and Immunochemical analysis of the reconstruction of DPLT. DPLTs (arrow) embedded in collagen gel a~d: hematoxylin and eosin (H&E) staining of the cells.

structed DPLT formed in groups that contained the added growth factor were more globular (Fig. 5). Hematoxylin and eosin (H&E) staining revealed that the size and compact structure of reconstructed DPLTs were very similar to actual DP (Figs. 6A, 6C, 6E, and 6G). We also evaluated the expression of versican, which is a well known marker of hair inducing activity and found that it was expressed strongly in the reconstructed DPLTs (Fig. 6). A compact structure was also observed for DPLTs without cell necrosis. Moreover, we detected the expression of type 4 collagen, laminin, and versican in reconstructed DPLTs employing MSCs in all groups. Because the DP volume controlled the thickness of the hair shaft, the DPLTs that were treated with NGF were unsuitable when compared to those that were treated with EGF and HGF [38]. The results of the immunohistochemical analysis revealed that EGF, HGF, and NGF were suitable for the reconstruction of DPLTs. Dermal papilla cells (DPCs) with hair-inducing activity tend to aggregate. The type 4 collagen and laminin expression of hUC-MSCs in the differ-


Fig. 7. The number of reconstructed DPLTs formed during culture in the presence of different growth factors for 6 days. (A) DPLTs in DPFM without growth factor as a control group, (B) DPLTs in the DPFM with 10 ng/mL EGF, (C) DPLTs in the DPFM with 20 ng/mL HGF, and (D) DPLTs in the DPFM with 10 ng/mL NGF.

Fig. 9. Western blotting of self-aggregated DPLTs. Cellular proteins (20 µg per lane) were separated on duplicate 15% polyacrylamide gels and then transferred onto nitrocellulose membranes. Each membrane was reacted with antiversican, anti-laminin antibody, and anti-beta actin antibody as a loading control. Lane 1, hUC-MSCs in DMEM/ 10% FBS; lane 2, DPFM supplemented with 3.2 uM HC; lane 3, DPFM supplemented with 3.2 uM HC and 10 ng/mL EGF; lane 4, DPFM supplemented with 3.2 uM HC and 20 ng/mL HGF; and lane 5, DPFM supplemented with 3.2 uM HC and 10 ng/mL NGF.

Table 2. The mean diameter and standard deviation of dermal papilla like tissue in response to treatment with various growth factors during self-aggregation Average diameter of Standard deviation DPLT (µm) (µm)

Fig. 8. RT-PCR of reconstructed DPLTs. Lane 1, Monolayer cultured hUC-MSCs in the proliferation medium; lane 2, DPLTs in the DPFM without growth factor; lane 3, DPLTs in the DPFM with 10 ng/mL EGF; lane 4, DPLTs in the DPFM with 20 ng/mL HGF; and lane 5 DPLTs in the DPFM with 10 ng/mL NGF.

Table 1. The mean diameter and standard deviation of dermal papilla like tissue produced using various initial inoculation cell densities Condition

Average diameter of DPLT (µm)

Standard deviation (µm)

4

2

98.635

38.572

4

1.6 × 10 cells/cm

2

157.857

62.352

4

2

2.4 × 10 cells/cm

160.505

85.278

4

2

172.362

75.356

0.8 × 10 cells/cm

4.0 × 10 cells/cm

DMEM/10% FBS

No formation

No formation

DPFM

92.978

47.471

DPFM + EGF10 ng/mL

167.257

55.620

DPFM + HGF 20 ng/mL

160.505

85.278

DPFM + NGF 10 ng/mL

132.725

55.553

entiation media was stronger than that of hUC-MSCs in the proliferation media. Additionally, all DPLTs expressed versican, which was reflected by their hair-inducing activity. We confirmed that preconditioning MSCs resulted in hairinducing activity through RT-PCR analysis (Fig. 8) and western blotting (Fig. 9). The number of reconstructed DPLT in DPFM containing EGF was highest in the 20 ng/ mL HGF group, followed by the 10 ng/mL EGF, and the 10 ng/mL NGF (Fig. 7) groups. When the cells were treated with NGF, cell aggregation was detected within 3 days of detach-attach step and the diameter of DPLTs was smaller when compared with the other groups (Table 2). RT-PCR analysis revealed that Type 4 collagen and laminin expression by hUC-MSCs in the differentiation media was stronger than the expression by hUC-MSCs in the proliferation media. All DPLTs expressed versican, which was reflected by their hair-inducing activity. Additionally, Type 4 collagen and laminin expression increased during preconditioning independently of the added growth factor

188

(Fig. 8). However, the morphogenic specific marker, versican was expressed strongly in cells that were treated with EGF or HGF (Fig. 9). These findings indicate that HGF could be substituted for EGF, which would reduce costs.

CONCLUSION The most effective treatment of alopecia developed to date is auto hair transplantation. The transplantation of human hair involves taking plugs of natural hair from areas in which occipital hair is growing and transplanting them to bald areas. However, the amount of hair that can be transplanted is severely limited, with the transplantation of only 2,000 hairs per operation and three operations being possible. Recently, studies have been conducted to evaluate the multiplication and transplantation of hair cells into foreign regions using cell isolation techniques and tissue engineering. Jahoda and Reynolds isolated rat vibrissae DP and transplanted cultivated DPCs via 3 passages in rat wounded ear skin and backs. After 4 to 8 weeks, they observed vibrissae type fibers in the back [41]. Reynolds et al. tested the inductive and immunoreactive properties of the human hair follicle dermis [42]. Recently, studies have been conducted to evaluate the multiplication and transplantation of hair cells into foreign regions using cell isolation techniques and tissue engineering. Jahoda and Reynolds isolated rat vibrissae dermal papilla (DP), and transplanted cultivated DP cells under 3 passages in rat wounded ear skin and backs. After 4 to 8 weeks, they observed vibrissae type fibers in the back [41]. Moreover, versican green fluorescent protein (GFP)-tagged transgenic mice exhibited GFP fluorescence in the DP cells of their pelage hair follicles and the DP cell fraction isolated by a high speed cell sorter showed hair inductivity when grafted with a newborn epidermis fraction [43]. Human hair can be regenerated by grafting hair follicles onto immunodeficient mice [44]; however, unlike murine cell-based hair regeneration, no hair follicle reconstitution from human cell preparations has been reported to date. Ehama et al. evaluated primary cultures of human foreskin derived epidermal cells that were co-grafted with murine DPCs and found that hair follicle-like structures were formed at the graft sites 4 weeks later. Hair shaft-like fibers developed with these keratinized structures and occasionally emerged at the skin surface. To determine if epithelial or mesenchymal cells affected or contributed to hair follicle formation, the ratio of each component was varied while fixing the other component at 1 × 107 cells. As the inoculation density of the dermal papilla cells was reduced, newly formed follicles diminished. These results indicate that the quantity of the dermal components is more critical than the epidermal components for follicle formation [45]. Jizeng et al. isolated follicular dermal and epidermal cells from embryonic mouse skin and found that it formed aggregates, which were subsequently cultured for 5~7 days, and then implanted intradermally into athymic mice. The mixed aggregates of murine follicular cells were found to have the ability to develop in culture into proto hairs that retained the ability to fully develop into hair folli-

cles after implantation [46]. Lin reported the production of reconstructed DP by enclosing DPCs within an alginatepolylysine-alginate (APA) semipermeable membrane. The microcapsules were implanted into rat footpads, which lack follicles and sebaceous glands, to assess their inductive properties. Histologic examination showed that follicles and sebaceous gland structures formed in the footpads within 6~10 weeks. At 10 weeks after transplantation, hair fibers were visible in the footpad. These findings indicate that the DP cell microcapsules retain the capacity to initiate follicle regeneration and could be considered as a substitute for freshly isolated DPs [47]. Osada developed a technique that involved aggregating cultured DP cells to form spheres in advance and then implanting them with epidermal cells. They were able to cultivate mouse DP cells for many passages by adding FGF-2. We found that utilizing sphere formation with cultured DP cells instead of using dissociated cells enhanced their follicle-inducing effects [20]. Ehama et al. evaluated primary cultures of human foreskin derived epidermal cells that were co-grafted with murine DPCs and found that hair follicle-like structures were formed at the graft sites 4 weeks later. Hair shaft-like fibers developed with these keratinized structures and occasionally emerged at the skin surface. Yoo et al. isolated and expanded umbilical cord-derived mesenchymal stem cells (UC-MSCs) from human Wharton’s Jelly. They found that culture-expanded UC-MSCs formed aggregates similar to native DP in size and morphology. Additionally, the aggregates showed a protein expression pattern similar to that of primary DPCs. To determine if the aggregates could induce hair follicles instead of DP, they were transplanted with ORSCs into the inner-dermis of athymic nude mice, where the cells underwent proliferation and differentiation. After 6 weeks of transplantation, they observed new hairs on the scalp of the mice [26]. This study was conducted to optimize the reconstruction step of dermal papilla like tissues. As in the case of MSCs, the dermal papilla like tissues required a differentiation step for dermal papilla cells to form prior to use in the cells. This generally requires the use of hepatocyte growth factor (HGF), which is relatively expensive. To reduce the cost of cell therapy using MSCs, we tested the effects of cell inoculation density and growth factor during differentiation. We found that EGF could be used to replace HGF during differentiation, which would reduce the cost of hair cell therapy. Received September 29, 2009; accepted November 13, 2009.

REFERENCES 1. Li, L. and R. M. Hoffman (1995) The feasibility of targeted selective gene therapy of the hair follicle. Nat. Med. 1: 705-706. 2. Yoo, B. Y., Y. H. Shin, H. H. Yoon, Y. J. Kim, K. Y. Song, S. J. Hwang, and J. K. Park (2007) Improved isolation of outer root sheath cells from human hair folli-


cles and their proliferation behavior under serum-free condition. Biotechnol. Bioproccess Eng. 12: 54-59. 3. Commo, S., O. Gaillard, and B. A. Bernard (2000) The human hair follicle contains two distinct K19 positive compartments in the outer root sheath: a unifying hypothesis for stem cell reservoir? Differentiation 66: 157164. 4. Seo, Y.-K., D.-H. Lee, Y.-H. Shin, B.-Y. Yoo, K.-M. Lee, K.-Y. Song, S.-J. Seo, S.-J. Hwang, Y.-J. Kim, E. -K. Yang, C.-S. Park, I.-S. Chang, and J.-K. Park (2003) Development of isolation and cultivation method for outer root sheath cells from human hair follicle and construction of bioartificial skin. Biotechnol. Bioproccess Eng. 8: 151-157. 5. McElwee, K. J., S. Kissling, E. Wenzel, A. Huth, and R. Hoffman (2003) Cultured peribulbar dermal sheath cells can induce hair follicle development and contribute to the dermal sheath and dermal papilla. J. Invest. Dermatol. 121: 1267-1275. 6. Elliott, K., T. J. Stephenson, and A. G. Messenger (1999) Differences in hair follicle dermal papilla volume are due to extracellular matrix volume and cell number: implications for the control of hair follicle size and androgen responses. J. Invest. Dermatol. 113: 873-877. 7. Katsuoka, K., C. Mauch, H. Schell, O. P. Hornstein, and Th. Krieg (1988) Collagen-type synthesis in human-hair papilla cells in culture. Arch. Dermatol. Res. 280: 140144. 8. Ozeki, M. and Y. Tabata (2003) In vivo promoted growth of mice hair follicles by the controlled release of growth factors. Biomaterials 24: 2387-2394. 9. Lee, Y. R. (2001) Hepatocyte growth factor (HGF) activator expressed in hair follicles is involved in in vitro HGF-dependent hair follicle elongation. J. Dermatol. Sci. 25: 156-163. 10. Shimaoka, S., R. Imai, and H. Ogawa (1994) Dermal papilla cells express hepatocyte growth factor. J. Dermatol. Sci. 7: S79-83. 11. Jindo, T., R. Tsuboi, R. Imai, K. Takamori, J. S. Rubin, and H. Ogawa (1995) The effect of hepatocyte growth factor/scatter factor on human hair follicle growth. J. Dermatol. Sci. 10: 229-232. 12. Yamazaki, M., R. Tsuboi, Y. R. Lee, K. Ishidoh, S. Mitsui, and H. Ogawa (1999) Hair cycle-dependent expression of hepatocyte growth factor (HGF) activator, other proteinases, and proteinase inhibitors correlates with the expression of HGF in rat hair follicles. J. Invest. Dermatol. Symp. Proc. 4: 312-315. 13. Peus, D. and M. R. Pittelkow (1996) Growth factors in hair organ development and the hair growth cycle. Dermatol. Clin. 14: 559-572. 14. Shimaoka, S., R. Tsuboi, T. Jindo, R. Imai, K. Takamori, J. S. Rubin, and H. Ogawa (1995) Hepatocyte growth factor/scatter factor expressed in follicular papilla cells stimulates human hair growth in vitro. J. Cell Physiol. 165: 333-338. 15. Jindo, T., R. Tsuboi, R. Imai, K. Takamori, J. S. Rubin, and H. Ogawa (1994) Hepatocyte growth factor/scatter

factor stimulates hair growth of mouse vibrissae in organ culture. J. Invest. Dermatol. 103: 306-309. 16. Jindo, T., R. Tsuboi, K. Takamori, and H. Ogawa (1998) Local injection of hepatocyte growth factor/scatter factor (HGF/SF) alters cyclic growth of murine hair follicles. J. Invest. Dermatol. 110: 338-342. 17. Katsuoka, K., H. Schell, B. Wessel, and O. P. Hornstein (1987) Effects of epidermal growth factor, fibroblast growth factor, minoxidil, and hydrocortisone on growth kinetics in human hair bulb papilla cells and root sheath fibroblasts cultured in vitro. Arch. Dermatol. Res. 279: 247-250. 18. Lu, Z. F., S. Q. Cai, J. J. Wu, and M. Zheng (2006) Biological characterization of cultured dermal papilla cells and hair follicle regeneration in vitro and in vivo. Chin. Med. J. (Engl) 119: 275-281. 19. Chiu, H. C., C. H. Chang, J. S. Chen, and S. H. Jee (1996) Human hair follicle dermal papilla cell, dermal sheath cell, and interstitial dermal fibroblast characteristics. J. Formos. Med. Assoc. 95: 667-674. 20. Osada, A., T. Iwabuchi, J. Kishimoto, T. S. Hamazaki, and H. Okochi (2007) Long-term culture of mouse vibrissal dermal papilla cells and de novo hair follicle induction. Tissue Eng. 13: 975-982. 21. Limat, A., T. Hunziker, E. R. Waelti, S. P. Inaebnit, U. Wiesmann, and L. R. Braathen (1993) Soluble factors from human hair papilla cells and dermal fibroblasts dramatically increase the clonal growth of outer root sheath cells. Arch. Dermatol. Res. 285: 205-210. 22. Yano, K., L. F. Brown, and M. Detmar (2001) Control of hair growth and follicle size by VEGF-mediated angiogenesis. J. Clin. Invest. 107: 409-417. 23. Yoo, H. G., I. Y. Chang, H. K. Pyo, Y. J. Kang, S. H. Lee, O. S. Kwon, K. H. Cho, H. Kwang, H. C. Eun, and K. H. Kim (2007) The additive effects of minoxidil and retinol on human hair growth in vitro. Biol. Pharm. Bull 30: 21-26. 24. Rho, S. S., S. J. Park, S. L. Hwang, M. H. Lee, C. D. Kim, I. H. Lee, S. Y. Chang, and M. J. Rang (2005) The hair growth promoting effect of Asiasari radix extract and its molecular regulation. J. Dermatol. Sci. 38: 89-97. 25. Charveron, M. and S. Lachgar (1998) Hair growth: VEGF (vascular endothelial growth factor) and fibroblasts of hair dermal papilla cells. Ann. Dermatol. Vener. 125: S9-S11. 26. Yoo, B. Y., Y. H. Shi, K. Y. Song, H. H. Yoon, and J. K. Park (2009) Evaluation of human umbilical cordderived mesenchymal stem cells on in vivo hair inducing activity. Tissue Eng. Regen. Med. 6: 15-22. 27. Method for the preparation of a dermal papilla tissue having hair follicle inductive potency. Pct kr 2005 004040: (2005) 28. Stenn, K. S. and R. Paus (2001) Controls of hair follicle cycling. Physiol. Rev. 81: 449-494. 29. Argyris, T. (1976) Kinetics of epidermal production during epidermal regeneration following abrasion in mice. Am. J. Pathol. 83: 329-340. 30. Ansel, J. C., C. A. Armstrong, I. Song, K. L. Quinlan, J.

190

31.

32.

33.

34.

35.

36.

37.

38.

39.

E. Olerud, S. W. Caughman, and N. W. Bunnett (1997) Interactions of the skin and nervous system. J. Investig. Dermatol. Symp. Proc. 2: 23-26. Arase, S., Y. Sadamoto, S. Katoh, Y. Urano, and K. Takeda (1990) Co-culture of human hair follicles and dermal papillae in a collagen matrix. J. Dermatol. 17: 667-676. Green, M. R. and J. R. Couchman (1984) Distribution of epidermal growth factor receptors in rat tissues during embryonic skin development, hair formation, and the adult hair growth cycle. J. Invest. Dermatol. 83: 118-123. du Cros, D. L., K. Isaacs, and G. P. Moore (1992) Localization of epidermal growth factor immunoreactivity in sheep skin during wool follicle development. J. Invest. Dermatol. 98: 109-115. Nanney, L. B., C. M. Stoscheck, L. E. Jr. King, R. A. Underwood, and K. A. Holbrook (1990) Immunolocalization of epidermal growth factor receptors in normal developing human skin. J. Invest. Dermatol. 94: 742748. Moore, G. P., B. A. Panaretto, and N. B. Carter (1985) Epidermal hyperplasia and wool follicle regression in sheep infused with epidermal growth factor. J. Invest. Dermatol. 84: 172-175. Donovan, M. J., R. Hahn, L. Tessarolloand, B. L. Hempstead (1996) Identification of an essential nonneuronal function of neurotrophin 3 in mammalian cardiac development. Nat. Genet. 14: 210-213. Ojeda, S. R., C. Romero, V. Tapia, and G. A. Dissen (2000) Neurotrophic and cell-cell dependent control of early follicular development. Mol.Cell Endocrinol. 163: 67-71. Thornton, M. J., K. Hamada, A. G. Messenger, V. A. Randall (1998) Androgen-dependent beard dermal papilla cells secrete autocrine growth factor(s) in response to testosterone unlike scalp cells. J. Invest. Dermatol. 111: 727-732. Chuang, Y. H., D. Dean, J. Allen, R. Dawber, and F. Wojnarowskaf (2003) Comparison between the expres-

sion of basement membrane zone antigens of human interfollicular epidermis and anagen hair follicle using indirect immunofluorescence. Br. J. Dermatol. 149: 274281. 40. Jahoda, C. A. (1992) Induction of follicle formation and hair growth by vibrissa dermal papillae implanted into rat ear wounds: vibrissa-type fibres are specified. Development 115: 1103-1109. 41. Reynolds, A. J. and C. A. Jahoda (1992) Cultured dermal papilla cells induce follicle formation and hair growth by transdifferentiation of an adult epidermis. Development 115: 587-593. 42. Reynolds, A. J., C. Lawrence, P. B. Cserhalmi-Friedman, A. M. Christiano, and C. A. B. Jahoda (1999) Transgender induction of hair follicles. Nature 402: 33-34. 43. Kishimoto, J., R. Ehama, L. Wu, S. Jiang, N. Jiang, and R. E. Burgeson (1999) Selective activation of the versican promoter by epithelial- mesenchymal interactions during hair follicle development. Proc. Natl. Acad. Sci. USA 96: 7336-7341. 44. Hashimoto, T., T. Kazama, M. Ito, K. Urano, Y. Katakai, N. Yamaguchi, and Y. Ueymamy (2000) Histologic and cell kinetic studies of hair loss and subsequent recovery process of human scalp hair follicles grafted onto severe combined immunodeficient mice. J. Invest. Dermatol. 115: 200-206. 45. Ehama, R., Y. Ishimatsu-Tsuji, S. Iriyama, R. Ideta, T. Soma, K. Yano, C. Kawasaki, S. Suzuki, Y. Shirakata, K. Hashimoto, and J. Kishimoto (2007) Hair follicle regeneration using grafted rodent and human cells. J. Invest. Dermatol. 127: 2106-2115 46. Qiao, J., A. Turetsky, P. Kemp, and J. Teumer (2008) Hair morphogenesis in vitro: formation of hair structures suitable for implantation. Regen. Med. 3: 683-692. 47. Lin, C. M., Y. C. Ji, H. Keng, X. N. Cai, and J. K. Zhang (2008) Microencapsulated human hair dermal papilla cells: a substitute for dermal papilla? Arch. Dermatol. Res. 300: 531-535.