Mechanical Stretch Upregulates SDF-1α in Skin Tissue and Induces Migration of Circulating Bone Marrow-Derived Stem Cells into the Expanded Skin


  • Shuang-Bai Zhou,

    1. Department of Plastic and Reconstructive Surgery, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Jing Wang,

    1. Department of Plastic and Reconstructive Surgery, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Cheng-An Chiang,

    1. Department of Plastic and Reconstructive Surgery, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Ling-Ling Sheng,

    1. Department of Plastic and Reconstructive Surgery, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Qing-Feng Li

    Corresponding author
    1. Department of Plastic and Reconstructive Surgery, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    • Correspondence: Qing-Feng Li, M.D., Ph.D., Department of Plastic and Reconstructive Surgery, Shanghai Ninth People's Hospital, 639 Zhizaoju Road, Shanghai 200011, China. Telephone: 86-21-23271699; Fax: 86-21-63089567; e-mail:

    Search for more papers by this author

  • S.Z.: conception and design, provision of study material or patients, data analysis and interpretation, and manuscript writing; J.W.: provision of study material or patients and collection and/or assembly of data; C.C. and L.S.: collection and/or assembly of data and data analysis and interpretation; Q.L.: conception and design, administrative support, and final approval of manuscript.


Background: Skin and soft tissue expansion is a procedure that stimulates skin regeneration by applying continuous mechanical stretching of normal donor skin for reconstruction purposes. We have reported that topical transplantation of bone marrow-derived mesenchymal stem cells (MSCs) can accelerate mechanical stretch induced skin regeneration. However, it is unclear how circulating MSCs respond to mechanical stretch in skin tissue. Methods: MSCs from luciferase-Tg Lewis rats were transplanted into a rat tissue expansion model and tracked in vivo by luminescence imaging. Expression levels of chemokines including macrophage inflammatory protein-1α, thymus and activation-regulated chemokine, secondary lymphoid tissue chemokine, cutaneous T-cell attracting chemokine, and stromal-derived factor-1α (SDF-1α) were elevated in mechanically stretched tissues, as were their related chemokine receptors in MSCs. Chemotactic assays were conducted in vitro and in vivo to assess the impact of chemokine expression on MSC migration. Results: MSC migration was observed in mechanically stretched skin. Mechanical stretching induced temporal upregulation of chemokine expression. Among all the tested chemokines, SDF-1α showed the most significant increase in stretched skin, suggesting a strong connection to migration of MSCs. The in vitro chemotactic assay showed that conditioned medium from mechanically stretched cells induced MSC migration, which could be blocked with the CXCR4 antagonist AMD3100, as effectively as medium containing 50 ng/ml rat recombinant SDF-1α. Results from in vivo study also showed that MSC migration to mechanically stretched skin was significantly blocked by AMD3100. Moreover, migrating MSCs expressed differentiation markers, suggesting a contribution of MSCs to skin regeneration through differentiation. Conclusion: Mechanical stretching can upregulate SDF-1α in skin and recruit circulating MSCs through the SDF-1α/CXCR4 pathway. Stem Cells 2013;31:2703–2713


Harvesting large skin tissue for reconstruction has always been a challenge for surgeons. Tissue expansion, developed by Neumann [1] and later improved by Radovan [2] and Austad et al. [3], is a procedure that stimulates and promotes skin regeneration through continuous mechanical stretching provided by an underlying silicone expander [4]. It has been used for a variety of plastic and reconstructive procedures due to its ability to supply donor skin with textures, structures, and adnexal distributions that closely match defect sites, without producing any secondary deformities at the donor sites [5]. However, in some cases, the skin can become too thin and undergo ulceration, even during routine expansion [6]. Thus, patients with a limited capacity for skin regeneration can hardly benefit from this technique. To improve the economic value and patient outcome of this technique, a new method must be developed to increase the expansion efficiency and shorten the expansion time.

Bone marrow-derived mesenchymal stem cells (MSCs) are multipotent stem cells obtained from bone marrow [7]. MSCs have the ability to promote tissue recovery and regeneration in various organs [8, 9]. We have reported that topical transplantation of MSCs can accelerate tissue expansion and skin regeneration [10]. However, the mechanism by which circulating MSCs respond to mechanical stretch in skin tissue and promote tissue regeneration is still unclear. Recently, it has been demonstrated that chemokines and their receptors are involved in recruiting MSCs [9]. We hypothesized that skin tissue undergoing mechanical stretch may synthesize and release a spectrum of cytokines that facilitate recruitment of circulating MSCs. In this study, we systemically transplanted MSCs into a tissue-expanded animal model to track the migration of MSCs in vivo. We also assessed the contribution of migrating MSCs in skin regeneration.

Materials and Methods

All animal procedures were conducted according to the Guide for the Care and Use of Laboratory Animals. All rats were maintained in a pathogen-free environment. All experiments were performed under laminar flow hoods.

Tissue Expansion Model

Wild-type female Lewis rats (4 weeks old, 75–84 g in weight) (Shanghai Experimental Animal Center, Shanghai, China) were anesthetized and 10-ml silicone expanders (Shanghai Xinsheng Biomedical Co., Ltd, Shanghai, China; were implanted into their backs subcutaneously [11]. Inflation was conducted every other day from 7 days postoperation. The controlled intracapsular pressure was 60 mmHg. The day at which inflation began was defined as Day 0.

Isolation and Culture of Luciferase-MSCs from Rat

Three-week-old luciferase (Luc)-Tg Lewis rats [12] (ROSA/Luciferase-LEW) (a generous gift from Prof. Eiji, Jichi Medical University, Jichi, Japan) were used as donors. Bone marrow cells were harvested by flushing femurs and tibiae with phosphate-buffered saline. These cells were filtered through 70-µm nylon mesh and plated in 100 mm culture dishes in Dulbecco's modified Eagle's medium (DMEM) (Gibco, Carlsbad, CA; supplemented with 10% fetal bovine serum (Hyclone, Waltham, MA; and 1% penicillin/streptomycin (Gibco, Carlsbad, CA; After 48 hours, the nonadherent cells were removed, and fresh medium was added to the dishes. The medium was then changed every 3 days, and the adherent spindle-shaped cells were further propagated for three passages.

Intravenous Transplantation of MSCs into Tissue Expansion Rats

Twelve female Lewis rats were randomized into two groups: an Expanded Group (A) and a Control Group (B), each consisting of six rats. Ten milliliter silicone expanders were implanted in rats from Group A and Group B according to the protocol described above. For Group A, routine inflation was conducted from 7 days postoperation. For group B, no additional inflation was performed.

Intravenous Luc-MSC transplantation was performed on Day 0, immediately before the inflation began. A total of 1 × 107 MSCs cultured from Luc-Tg rats (Luc-MSCs) were suspended in DMEM and injected intravenously through the retrobulbar venous plexus. All rats were observed for 3 weeks after cell transplantation.

In Vivo Bioluminescent Imaging

In vivo luminescent imaging was performed with a noninvasive bioimaging system IVIS (Xenogen, Alameda, CA; capable of continuously tracking the migration of implanted MSCs. MSC movement was analyzed using the IVIS Living Image (Xenogen, Alameda, CA; software package. To detect luminescence from luciferase-expressing cells, D-luciferin (potassium salt, Synchem, Elk Grove Village, IL; was injected intraperitoneally in rats anesthetized with isoflurane (30 mg/kg b.wt.). Signal intensity was quantified as photon flux in units of photons/sec/cm2/steradian (p/s/cm2/sr) in the region of interest. In vivo bioimaging was performed on Day 0, Day 1, Day 4, Day 7, Day 14, and Day 21. To identify the proportion of migrated cells, we took bioimage data from a group of expansion model rats immediately after intradermal injection with same quantity of Luc-MSCs, to evaluate the baseline of transplanted MSCs.

Immunofluorescence Staining

To track and detect the differentiation of transplanted MSCs, expanded skin sections were routinely fixed with paraformaldehyde and embedded in Tissue Tec OCT Compound (Fisher Scientific, Waltham, MA; Anti-firefly luciferase antibody (Abcam, Cambridge, UK; was applied to label the transplanted Luc-MSCs. In addition, skin sections were treated with primary antibodies against Keratin 19 (K19) (Millipore, Hayward, CA;, Keratin 15 (K15) (Santa Cruz, Dallas, TX;, and CD31 (Abcam, Cambridge, UK; Primary antibodies were visualized using secondary antibodies conjugated to either 488- or 555-nm absorbing fluorophores (Abcam, Cambridge, UK; Fluorescence staining was detected using a confocal laser scanning fluorescence microscope (Carl Zeiss, Jena, Germany;

Expression of Chemokines in Mechanically Stretched Skin

Tissue expansion models without transplanted MSCs were used for chemokine expression analysis. Five rats were sacrificed before expander inflation and at Day 1, Day 4, Day 7, Day 14 and Day 21, respectively. The expanded dorsal skin samples were harvested for chemokine spectrum examination. Another five rats without implanted expanders were sacrificed, and their dorsal skin was harvested for control samples.

Total RNA was isolated from normal or expanded skin tissues with Trizol reagent. Real-time quantitative polymerase chain reaction analysis (RT-qPCR) was performed with primers against the following chemokines: stromal-derived factor-1α (SDF-1α, CXCL12; sense: 5′-CTTTTCAGCCTTGCAACAATC-3′, antisense: 5′-TGCATCAGTGACGGTAAGCCA-3′), macrophage inflammatory protein-1α (MIP-1α, CCL3; sense: 5′-TCCCAGCCGGGTGTCATTTTCCT-3′, antisense: 5′-CTGCTCTACACGGGGCCCAC-3′), thymus and activation-regulated chemokine (TARC, CCL17; sense: 5′-ATGATGTCACTTCAGATGC-3′, antisense: 5′-GCACTCTCGGCCTACATTGG-3′), secondary lymphoid tissue chemokine (SLC, CCL21; sense: 5′-GCAAGGGGACTGAACAGACA-3′, antisense: 5′-ATGGAGAGCAAGTGCAGGTC-3′), and cutaneous T-cell attracting chemokine (CTACK, CCL27; sense: 5′-GAGCGGAGTCCGATGCCTC-3′, antisense: 5′-GCTTCTC CTTAGTCTTGTTCCA-3′). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; sense: 5′-GGGGCTCTCTGCTCCTCCCTGT-3′, antisense: 5′-CGGCCAAATCCGTTCACACCGA-3′) was applied as a control to verify RNA integrity. Relative gene expression was calculated as reported previously [13]. The target gene mRNA abundance relative to the control was calculated by the formula math formula. All assays were performed in duplicate. Immunohistochemical staining with antibody of SDF-1α (Abcam, Cambridge, UK; was applied to further analysis the SDF-1α expression in expanded skin.

Chemokine Receptors Expression in MSCs

Total RNA was isolated from MSCs. Reverse transcription-polymerase chain reaction (RT-PCR) was performed to analyze expression of chemokine receptors in MSCs with primers against CXCR4 (sense: 5′-ACTGCATCATCATCTCCAAGC-3′; antisense: 5′-CTCTCGAAGTCACATCCTTGC-3′), CCR4 (sense: 5′-TCTACAGC--GGCATCTTCTTCAT-3′, antisense: 5′-CAGTACGTGTGGTTGTGCTCT-3′), CCR5 (sense: 5′-CTGGCCATCTCTGACCTGTTTTTC-3′, antisense: 5′-CAGCCC--TGTGCCTCTTCTTCTCAT-3′), CCR7 (sense: 5′-ACAGCGGCCTCCAGAAGA-AGAGCGC-3′, antisense: 5′-TGACGTCATAGGCAATGTTGAGCTG-3′), CCR10 (sense: 5′-GGCCCTGACTTTGCCTTTTG-3′, antisense: 5′-GCTGCCAGTAGAT-CGGCTGT-3′, and GAPDH. The products were resolved by electrophoresis on a 1.5% agarose gel. Gels were imaged under UV light, and the relative band intensity was evaluated and normalized to GAPDH using WinCAM software (Cybertech, Berlin, Germany). Comparisons were performed by subjecting samples to amplification in subsaturating conditions with the constitutively expressed GAPDH gene and the chemokine receptors of interest in parallel tubes.

In Vitro Cell Migration Assays

Rat MSC migration was assessed in 24-well plates with 8 mm pore size polycarbonate transwell culture inserts (Corning, Corning, NY; MSCs were placed on top of transwell membranes in the upper chamber at a density of 1 × 105 cells/well. For migration inhibition experiments, MSCs were preincubated for 30 minutes with 5 µg/ml AMD3100, a CXCR4 antagonist [14]. To observe the chemotactic effect of SDF-1α on MSC migration, cell culture medium alone (DMEM, GIBCO, NY) or medium containing 50 or 100 ng/ml rat recombinant SDF-1α (rrSDF-1α) (Peprotech, Rocky Hill, NJ; was added to individual lower chambers. Moreover, conditional medium (CM) from mechanical stretch-treated rat keratinocytes (harvested by routine protocols) was applied for evaluating mechanical stretch induced SDF-1α secretion. Static mechanical tension (SMT) at 10% elongation was applied for 12 hours using an FX-4000T Flexcell Tension Plus unit (Flexcell, Hillsborough, NC; The mechanical stretch-treated CM (SMT-CM) was collected immediately after completion of SMT stimulation. Control CM was collected from keratinocytes cultured on the same plates in the same incubator for the same time that were not subjected to tension.

After 6 hours of incubation, the membrane of transwell inserts was removed and fixed after stripping to detach nonmigrating cells. The number of MSCs that migrated to the lower surface of the membrane was counted for five random high-power fields (HPFs, 200×) from microscopic images (Olympus IX70, Olympus, Tokyo, Japan; Experiments were performed in triplicate. Data were expressed as mean number of cells per HPF ± SD.

In Vivo MSC Migration Inhibition Assay

To assess the connection between the SDF-1α/CXCR4 axis and MSC migration into expanded skin, an in vivo MSCs migration inhibition assay was performed. Eighteen rats implanted with expanders were divided into three groups. In the MSC Group and the Control Group, 1 × 107 Luc-MSCs were injected as described above. In the AMD-MSCs Group, Luc-MSCs were incubated with 5 µg/ml AMD3100 for 30 minutes before injection. The rats from MSCs Group and AMD-MSCs Group received regular expander inflation. Luc-MSC migration was observed continuously for 3 weeks.

Histological Examination of Expanded Skin

Whole area of expanded skin tissues from MSC Group, AMD-MSC Group, and Control Group were collected and compared for the expanded skin area at the end of 3 weeks follow-up. After routinely fixed with paraformaldehyde, sections of 7 µm were prepared and stained with hematoxylin and eosin for expanded skin thickness evaluation. Immunohistochemical staining with anti-proliferating cell nuclear antigen (PCNA) antibody (Abcam; was used for detection of proliferating cells. The number of PCNA positive cells was counted for five random HPFs (200×) from microscopic images (Olympus IX70, Olympus, Tokyo, Japan).


Results are expressed as mean ± SD. Statistical differences were determined by a two-tailed Student's t test. p-Values less than .05 were considered statistically significant.


Characteristics of MSCs Isolated from Luciferase Transgenic Rats

MSCs were isolated from Luc-Tg rats and cultured ex vivo. We confirmed luciferase expression by in vitro luciferase bioimaging. MSCs from Luc-Tg rats were plated onto 96-well plates, and luciferase activity was confirmed in the presence of D-luciferin (Fig. 1A).

Figure 1.

Transplantation of Luc-mesenchymal stem cells (MSCs) into tissue expansion models. (A): Luciferase activity of MSCs from Luc-Tg rats. Representative luciferase image taken using IVIS (Xenogen, Alameda, CA). (B, C): In vivo luciferase imaging of systemic transplanted Luc-MSCs. A total of 1 × 107 Luc-MSCs were injected into the retrobulbar venous plexus. The Luc-MSC signal was monitored with the IVIS system (Xenogen). The signal decreased greatly on Day 1 but increased beginning at Day 4. The peak Luc-BMSCs signal was observed on Day 7 and gradually faded on Day 14.

In Vivo Detection of MSCs Migrating to the Expanded Skin

To investigate whether circulating MSCs migrated into the expanded skin, we transplanted MSCs transvenously into tissue expansion rats (Group A) and normal rats (Group B). The results of in vivo bioimaging showed that, in both groups, photons emitted from the Luc-MSCs were detected in the lungs and spleen on Day 1, and occasionally in the orbital region. In Group A, the signal in the lungs and spleen gradually began to fade as the signal began to appear in the expanded dorsal skin on Day 4. The Luc-MSC signal peaked in the expanded skin on Day 7 and faded on Day 14 and Day 21 (Fig. 1B, 1C). Whereas, in Group B, Luc-MSC signal did not appear gathering in dorsal skin (Supporting Information Fig. 1A). When compared with total amount of transplanted MSCs, approximately 19.42% of transplanted MSCs at peak were gathered to expanded skin on Day 7 in Group A (Supporting Information Fig. 1B, 1C).

Chemokines Expression in Normal and Expanded Skin

The expression of MIP-1α, TARC, SLC, CTACK, and SDF-1α in the expanded skin flap was evaluated by qRT-PCR (Fig. 2A, 2B). The levels of several chemokines increased over time. Among all the tested chemokines, SDF-1α showed the most significant upregulation. The expression of SDF-1α began to increase significantly from Day 1 (18.73 ± 2.99-fold over the control, p < .01) and reached its peak on Day 7 (67.65 ± 8.55-fold over the control, p < .01). From Day 14 to Day 21, the level of SDF-1α gradually decreased (21.56 ± 2.46-fold over the control). TARC expression peaked at Day 7 (8.94 ± 1.61-fold over Day 0), while SLC and CTACK reached their peaks at Day 14 (5.88 ± 1.27 and 4.75 ± 2.73-fold over the control, respectively). MIP-1α levels fluctuated from 3.54 ± 1.79 to 6.95 ± 1.90 times that of the control between Day 4 and Day 21. The expression of SDF-1α in expanded skin was upregulated by mechanical stretch more than the other chemokines were.

Figure 2.

Expression of the chemokine spectrum in mechanically stretched skin. (A, B): Real-time PCR analysis of mechanically stretched skin. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was applied as an internal control. A significant upregulation of SDF-1α occurred on Day 1 and reached its peak on Day 7. Expression of MIP-1α, TARC, SLC, and CTACK also had varying degrees of upregulation from Day 1 to Day 21. (C): Immunohistochemical staining with an anti-SDF-1α antibody in mechanically stretched skin. Mechanically stretched skin harvested on Day 7 and Day 14 had significantly more SDF-1α-positive cells than the control, especially in the epidermis and in hair follicles (*, p < .05). Bars = 50 µm. (D, E): Expression of chemokine receptors in mesenchymal stem cells (MSCs). GAPDH was applied as an internal control. CXCR4, CCR5, and CCR10 were all found in cultured MSCs, while CCR4 and CCR7 were both expressed at low levels. Abbreviations: CTACK, cutaneous T-cell attracting chemokine; MIP-1α, macrophage inflammatory protein-1α; SDF-1α, stromal-derived factor-1α; SLC, secondary lymphoid tissue chemokine; TARC, thymus and activation-regulated chemokine.

We next evaluated the pattern of SDF-1α expression in expanded tissue using immunohistochemical staining. Mechanically stretched skin harvested on Day 7 and Day 14 had significantly higher SDF-1α-positive cells than the control, especially in epidermis and hair follicles (Fig. 2C).

Chemokine Receptor Expression in MSCs

RT-PCR was performed to investigate chemokine receptor expression in MSCs. MSCs cultured to passage 4 were used for expression analysis of CXCR4, CCR4, CCR7, CCR10, and CCR5, which are specific receptors for SDF-1α, TARC, SLC, CTACK, and MIP-1α, respectively. The RT-PCR analysis indicated that MSCs expressed CXCR4, CCR5, CCR7, and CCR10. However, CCR4 was not expressed in MSCs (Fig. 2D). The upregulation of SDF-1α in the expanded skin along with the high expression of its receptor CXCR4 in MSCs suggests that MSCs were recruited to the expanded skin by the SDF-1α/CXCR4 pathway.

Migration of MSCs in Response to SDF-1 α In Vitro

A dose-dependent chemotactic effect of rrSDF-1α was found toward MSCs in vitro. The optimal effect was observed at 50 ng/ml and 100 ng/ml, but the chemotactic effect of rrSDF-1α was abolished when MSCs were preincubated with AMD3100 (p < .01) (Fig. 3A--3D). SMT-MC had a similar effect as that of 50 ng/ml rrSDF-1α on MSC migration (Fig. 3E). When MSCs were pretreated with AMD3100, the chemotactic effect of SMT-CM was greatly decreased (Fig. 3F). Control-CM did not stimulate a chemotactic response (Fig. 3G).

Figure 3.

In vitro chemotactic assay for mesenchymal stem cells (MSCs). Culture medium with different concentrations of recombinant rrSDF-1α protein (0 ng/ml (A), 50 ng/ml (B), and 100 ng/ml (C)) was applied to measure MSC migration induced by SDF-1α, which could be blocked by AMD3100 pretreatment (D). SMT-CM was collected after MSCs were stretched for 12 hours. In the chemotactic assay, SMT-CM could induce MSC migration with a similar effect to that of 50 ng/ml rrSDF-1α (E). When MSCs were pretreated with AMD3100, migration was greatly decreased (F). Control-CM was similar to DMED without rrSDF-1α (G). AMD3100 could block the stimulating effect of 50 ng/ml SDF-1α and SMT-CM (H) (*, p < .05). Bars = 100 µm. Abbreviations: HPF, high-power field; SDF, stromal-derived factor.

Inhibition of the SDF-1α/CXCR4 Pathway Reduces Migration of MSCs Toward Expanded Skin

To further verify the chemotactic effect of the SDF-1α/CXCR4 pathway in MSC recruitment to expanded skin, MSCs were pretreated with the CXCR4 antagonist AMD3100 before intravenous transplantation. In vivo bioimaging revealed that the luciferase signal from the MSC Group increased significantly up to Day 7 (2,110 ± 178 p/s/cm2/sr), reaching its peak at Day 11 (2,310 ± 220 p/s/cm2/sr), before decreasing slightly (Fig. 4A). However, in the AMD-MSC Group, MSCs did not aggregate in the expanded skin (Fig. 4B) and showed no difference as indicated by luciferase signal from the Control Group (p = .11) (Fig. 4C). This suggests that AMD3100 blocks the chemotactic response of MSCs induced by mechanical stretch (Fig. 4D).

Figure 4.

In vivo bioimaging for the stromal-derived factor-1α/CXCR4 pathway induced Luc-MSCs migration. Silicone expanders were implanted in the backs of rats. Regular saline injections were administered in the MSC Group and the AMD-MSC Group but not in the Control Group. Luc-MSCs were transplanted via the retrobulbar venous plexus in rats from the MSC Group and the Control Group, while rats from the AMD-MSC Group were treated with AMD-3100 pretreated Luc-MSCs (AMD-3100-MSCs). The Luc-MSC signal was observed in mechanically stretched skin on Day 7 in the MSC Group (A). MSC migration was not observed in either the AMD-MSC Group (B) or the Control Group (C). The time course of luminescence is depicted. The MSC Group had significantly higher luminescence in the expanded skin than the AMD-MSC Group and the Control Group (D) (*, p < .05). Abbreviations: MSC, mesenchymal stem cell.

Immunohistochemical staining of skin sections from each group showed similar results. Skin sections from the MSC Group contained more luciferase+ cells than the AMD-MSC Group on Day 7 (207 ± 17 vs. 34 ± 10 cells/HP, p < .001) and on Day 14 (252 ± 53 vs. 52 ± 17 cells/HP, p < .001). However, when compared with the Control Group, more luciferase+ cells were detected in the AMD-MSC Group both on Day 7 (34 ± 10 vs. 18 ± 5 cells/HP, p < .01) and on Day 14 (52 ± 17 vs. 16 ± 5 cells/HP, p < .001) (Fig. 5). These results indicate that while the SDF-1α/CXCR4 interaction plays an important role in migration of MSCs toward expanded skin, it is not the only pathway involved in recruiting MSCs.

Figure 5.

In vitro tracking of transplanted Luc-MSCs. Mechanically stretched skin tissue was collected on Day 7 (A) and Day 14 (B) from three groups and fixed with paraformaldehyde. Tissue sections were stained with anti-luciferase antibody for tracking the transplanted MSCs. When compared with the AMD-MSCs Group and the Control Group, skin tissue from the MSCs Group contained more luciferase+ cells on Day 7 (C) and on Day 14 (D). (*, p < .05) Bars = 50 µm. Abbreviations: HPF, high-power field; MSC, mesenchymal stem cell.

MSCs Recruited by SDF-1α/CXCR4 Promote Expanded Skin Regeneration

To evaluate migrated MSCs contribution to expanded skin regeneration, we evaluated expanded skin from MSC Group and AMD-MSC Group by area, thickness of epidermis, and proliferating cell number. When checking the area of expanded skin on Day 14, skin from MSC Group had modest but significant superiority to AMD-MSC Group (31.57 ± 2.51 cm2 vs. 21.42 ± 2.23 cm2, p < .01) (Fig. 6A, 6B). Expanded epidermal thickness was measured after H&E staining. Thickness of skin from MSC Group was 131.76 ± 21.78 µm while that from AMD-MSC Group was 78.44 ± 14.38 µm (p < .01) (Fig. 6C, 6D). We also use anti-PCNA primary antibody to mark proliferating cells in expanded skins. There were more PCNA+ cells observed in skin section from MSC Group (30 ± 5 cells/HP) than AMD-MSC Group (16 ± 3 cells/HP, p < .01) (Fig. 6E, 6F). These results showed that expanded skin gained better proliferation and regeneration capacity with intravenously MSCs transplantation.

Figure 6.

MSCs recruited by stromal-derived factor-1α/CXCR4 pathway promote expanded skin regeneration. Larger expanded skin tissue was harvested in MSC-Group than in AMD-MSC Group after 3 weeks expansion (A, B). Epidermal thickness was analyzed after HE staining. Expanded skin from MSC Group had significant advantage in epidermal thickness than that from AMD-MSC Group (C, D). By marking proliferating cells with anti-PCNA antibody, the results showed that there were more PCNA+ cells in skin sections from MSC Group than those from AMD-MSC Group (E, F) (*, p < .05). Bars = 50 µm. Abbreviations: HPF, high-power field; MSC, mesenchymal stem cell.

MSCs Contribute to Skin Regeneration by Differentiating to Multiple Cell Types

An anti-luciferase antibody was used to locate transplanted MSCs in the expanded skin tissue. Luciferase+ cells were mainly found in the epidermis, dermis, interfollicular epidermis, and hair bulbs (Fig. 7). We also stained with anti-K15, anti-K19, and anti-CD31 antibodies to examine MSC differentiation. Cells expressing both luciferase and K19, as well as those expressing both luciferase and K15, represented epidermal stem cells. Differentiated epidermal stem cells were located in the upper portions of the hair follicles, interfollicular epidermis, and sebaceous glands (Fig. 7). MSCs also differentiated into endothelial cells, as evidenced by luciferase+/CD31 cells located in blood vessels (Fig. 7). These results demonstrated that migrating MSCs contributed to skin regeneration by differentiating into multiple cell types.

Figure 7.

Detection of transplanted mesenchymal stem cell (MSC) differentiation in mechanically stretched skin. Skin sections were stained with an anti-luciferase antibody to label transplanted MSCs, as well as anti-K19 antibody (A--D), anti-K15 antibody (E--H), and anti-CD31 (I--L). Some of the migrated MSCs differentiated into epidermal stem cells and endothelial cells. Bars = 50 µm.


In this study, we reported an upregulation of chemokines in mechanically stretched skin. When undergoing mechanical stretch, skin tissue displayed a time-dependent change in chemokine expression. We also found that circulating MSCs could respond to mechanical stretch and migrate to expanded skin. Our results suggest that SDF-1α, one of most significantly upregulated chemokines, plays an important role in mediating MSC migration into stretched skin. Our findings also indicate that MSC could contribute to expanded skin regeneration by transdifferentiating into several cell types in the expanded skin.

Tissue repair and regeneration involve selective recruitment of circulating or resident stem cell populations. MSCs have been investigated for use in organ regeneration because of their ability to undergo multipotent differentiation. MSCs can also selectively migrate to injured sites, including bone fractures [15, 16], myocardial infarctions [17-19] and ischemic cerebral injuries [9], where they contribute to tissue recovery. This has also been demonstrated for skin wound healing [20-22].

Skin and soft tissue expansion is a method for in situ tissue regeneration that involves complex interactions between cells and cytokines [23]. MSC participation in skin expansion and regeneration is seldomly discussed, although previous studies demonstrated that intradermal transplantation of MSCs can promote regeneration [10]. In our study, we proved that circulating MSCs can be recruited to and migrate into expanded skin. In our in vivo bioimaging study, there were approximately 19.42% of systemically transplanted MSCs migrated into the expanded skin following expander inflation, despite the fact that major portion of MSCs were captured by the lung and spleen within 48 hours of cell injection [24]. The number of migrated MSCs reached its peak at Day 7 and slightly decreased from Day 14 to Day 21.

Selective migration and engraftment of implanted MSCs may involve a spectrum of cytokines released by target tissues. Interactions between chemokines and their specific receptors are expected to contribute to migration and engraftment [9]. Several chemokines are upregulated in injured tissues and have been verified to recruit MSCs expressing the appropriate chemokine receptors [19]. For the first time, we investigated changes in chemokine expression in mechanically stretched skin. We tested several chemokines including MIP-1α, TARC, SLC, CTACK, and SDF-1α, which have all been reported to be involved in tissue recovery [15, 17, 22, 25-27]. Although multiple chemokines were upregulated at the RNA level, SDF-1α had the most significant augmentation in expression. The expression of SDF-1α increased starting on Day 1 and reached its peak on Day 7, before gradually decreasing. CXCR4, the receptor for SDF-1α, was highly expressed in MSCs. Therefore, we presumed that the SDF-1α/CXCR4 pathway was important for recruiting MSCs to expanded skin tissue.

The SDF-1α/CXCR4 axis contributes to MSCs migration in multiple tissues [15, 17, 27-29]. The results of our in vivo bioimaging revealed that, when pretreated MSCs with AMD3100, a significant decrease in the number of luciferase+ cells was observed migrating to expanded skin in the AMD-MSC Group than in the MSC Group starting from Day 1. The results of immunofluorescent staining also showed that fewer luciferase+ cells were found in the AMD-MSC Group, both on Day 7 and Day 14. This confirmed that recruitment of MSCs to expanded skin depends on SDF-1α/CXCR4 interaction.

An in vitro chemotactic assay was performed to confirm that the SDF-1α/CXCR4 axis was involved in MSC migration induced by mechanical stretch. CM from stretch-treated keratinocytes was also used in the chemotactic assay. CM performed as well as the 50 ng/ml rrSDF-1α solution at inducing MSC migration and this effect could also be blocked with AMD3100. These results strongly imply that mechanical stretch can upregulate SDF-1α expression in skin tissue and induce MSC migration.

The upregulation of SDF-1α following mechanical stretching of skin may arise from induction of hypoxia inducible factor-1α (HIF-1α). Increased expression of HIF-1α in response to mechanical stretching has been reported for vascular smooth muscle cells [30, 31], myocardial cells [32], skeletal muscle fibers [33, 34], and fibroblasts [35]. Mechanical stretching can be sensed by cell membrane ion channels, integrins, and receptor tyrosine kinases [36]. It can also induce activation of HIF-1α via PI3K/Akt/mTOR [34, 37] or MAPK pathways [30], which may lead to increased SDF-1α expression. In addition, cyclic stretch has been reported to upregulate SDF-1α expression in human smooth muscle cells [38].

During our research, we noticed that there were still small amount of MSCs migrated to expanded skin after blocking SDF-1α/CXCR4 axis with AMD3100. That may be caused of either the interaction of SDF-1α/CXCR7 axis or other upregulated chemokines in expanded skin. Even though, results from our researches indicate that SDF-1α/CXCR4 pathway has the key role in recruiting MSCs and promoting expanded skin regeneration. The results showed that, after block SDF-1α/CXCR4 pathway by AMD3100, the number of proliferating cells in expanded skin was significantly decreased (30 ± 5 cells/HP vs. 16 ± 3 cells/HP, p < .01). The results also demonstrated that larger and thicker expanded skin was gained in MSC Group than AMD-MSC Group (p < .01).

There is still controversy about the mechanism of MSC recruitment and acceleration of tissue regeneration. It is believed that MSCs promote tissue regeneration by two pathways: growth factor paracrine and cell differentiation [8, 9]. We have previously reported that MSCs may promote expanded skin regeneration through paracrine effects [10]. In this study, we demonstrated that migrating MSCs supported tissue regeneration by differentiation. Luciferase+ MSCs were located in the upper portions of hair follicles, interfollicular epidermis, and sebaceous glands as well as sporadically in the dermis. Immunofluorescence staining results showed that some of these luciferase+ cells expressed epidermal stem cell markers such as K19 and K15 [39-41], indicating that engrafted MSCs directly contribute to skin regeneration via differentiation and proliferation. Our findings are in accordance with other research on wound healing. Moreover, luciferase+/CD31+ MSCs were also observed, indicating that MSC transplantation might ensure better blood supply for expanded skin by forming blood vessels.

Recently, it has been speculated that specific microenvironments surrounding MSCs, also known as niches, lead to MSC transdifferentiation into organ-specific cells [8, 9]. Thus, accelerating MSC migration to specific sites may be important for developing MSC-based treatments. Our findings indicate that chemokines may be effective at improving MSC engraftment. Although our research was focused on the SDF-1α/CXCR4 axis, we believe that there are other pathways involved in MSC migration to expanded skin.


In this article, we showed that the expression of several chemokines changed over time in mechanically stretched skin. Among these chemokines, SDF-1a displayed the most significant upregulation and was effective at promoting MSC migration into expanded skin. Our results also confirmed that engrafted MSCs contributed to skin regeneration by directly differentiating into various cell types.


This research was supported by the Key Project of National Natural Science Foundation No. 30730092 and National Science Fund for Distinguished Young Scholars No. 30925034. We would like to thank Dr. Eiji Kobayashi, Jichi Medical University, Japan, for donating luciferase-transgenic rats as a generous gift. We would also like to thank Dr. Mei Yang for language polishing and professional editing.

Disclosure of Potential Conflicts of Interest

We indicate no potential conflicts of interest.