Chinese Medical Journal 2005;118(3):186-191
Morphological and distribution characteristics of sweat glands in hypertrophic scar and their possible effects on sweat gland regeneration

FU Xiao-bing,  SUN Tong-zhu,  LI Xiao-kun,  SHENG Zhi-yong

FU Xiao-bing (Laboratory of Wound Healing and Cell Biology, Burns Institute, 304th Clinical Department, General Hospital of PLA , Beijing 100037, China)

SUN Tong-zhu (Laboratory of Wound Healing and Cell Biology, Burns Institute, 304th Clinical Department, General Hospital of PLA , Beijing 100037, China)

LI Xiao-kun (Biological Research Laboratory, Jinan University, Guangzhou 510632, China)

SHENG Zhi-yong (Laboratory of Wound Healing and Cell Biology, Burns Institute, 304th Clinical Department, General Hospital of PLA , Beijing 100037, China)

Correspondence to:FU Xiao-bing,Laboratory of Wound Healing and Cell Biology, Burns Institute, 304th Clinical Department, General Hospital of PLA, Beijing 100037, China (Tel: 86-10-66867396. Fax:86-10-88416390.;
burns;cicatrix, hypertrophic;sweat gland;regeneration
Background In hypertrophic scar tissue, no sweet gland and hair follicle exist usually because of the dermal and epidermal damage in extensive thermal skin injury, thus imparing regulation of body temperature. This study was designed to reveal the morphological and distributional characteristics of the sweat glands in normal skin and hypertrophic scar obtained from children and adults, and to study the possible interfering effects of the scar on regeneration of the sweat gland after burn injury.
Methods Biopsies of hypertrophic scar were taken from four children (4-10 years) and four adults (35-51 years). Normal, uninjured full-thickness skin adjacent to the scar of each patient was used as control. Keratin 19 (K19) was used as the marker for epidermal stem cells and secretory portion of the sweat glands, and keratin 14 (K14) for the tube portion, respectively. Immunohistochemical and histological evaluations were performed.
Results Histological and immunohistochemical staining of skin tissue sections from both the children and adults showed K19 positive cells in the basement membrane of epidermis of normal skin. These cells were seen only single layer and arranged regularly. The secretory or duct portion of the eccrine sweat glands was situated in the dermis and epidermal layer. However, in the scar tissue, K19 positive cells were scant in the basal layer, and the anatomic location of the secretory portion of sweat glands changed. They were located between the border of the scar and reticular layer of the dermis. These secretory portions of sweat glands were expanded and were organized irregularly. But a few K14 positive cells were scattered in the scar tissues in cyclic form.
Conclusions There are some residual sweat glands in scar tissues, in which the regeneration process of active sweat glands is present. Possibly the sweat glands could regenerate from adult epidermal stem cells or residual sweat glands in the wound bed after burn injury.
Wound healing is a complex biological process, which involves bleeding, inflammatory reaction, granulation and re-epithelialization. Hopeful wound healing should include the re-establishment of both skin anatomic structure and its function.[1,2] In many wounds, however, hypertrophic scar (HSc) formation is a common clinical problem for patients who survive extensive thermal injury.[3,4]

Histologically, hypertrophic scar is an excessive accumulation of extracellular matrix (ECM) proteins, resulting in an imbalance between the synthesis and degradation of protein. The presence of nodules of thin and densely packed collagen fibers in the dermis is the most characteristic feature of this fibroproliferative condition. Usually, there are no sweet glands and hair follicles in hypertrophic scar tissue because of the dermal and epidermal damage caused by extensive thermal skin injury, which impairs the regulation of body tem-perature.[3,4]

In normal skin, there are two types of sweat glands, apocrine and eccrine. The apocrine sweat gland found primarily in the axillae and public region consists of a coiled secretory gland in the lower portion of the dermis and a straight excretory duct that empties onto the infundibular component of the hair follicle, which is normally adjacent to the sebaceous gland. As the main regulator of body temperature, several million eccrine sweat glands, weighing about 100 g in an estimated aggregate mass are distributed all over the body surface.[5] These glands comprised a coiled gland and three sections of an eccrine duct. The coiled dermal duct with a coiled gland lies in the dermis and gives rise to a straight duct, which connects with the spiral duct, located in the epidermis and the spiral exit to the skin surface. These sweat glands can lower body temperature by producing a film of water on the surface of the skin that yields evaporative cooling. Traditionally, unlike the epidermis and the hair follicle, eccrine glands do not regularly renew themselves via cell division and terminal differentiation.[6]

A number of hypotheses, including embryonic development and histology, have been proposed that sweat glands may regenerate from adult epidermal stem cells or remain in the wound bed. However, little is known about the morphological and distribution characteristics of sweat glands in scar tissue and the relationship between epidermal stem cells and sweat gland development. In this study, we explored the morphological and distribution characteristics of sweat glands and their relations to epidermal stem cells in human skin. Also, we studied the distribution characteristics of secretory portion and duct portion of the eccrine sweat glands in hypertrophic scar tissue. These efforts were made to offer direct evidences for inducing sweat gland regeneration from adult epidermal stem cells or residual sweat glands in injured skin.


Skin biopsies
Eight patients aged from 4 to 10 years (young group, n=4) and 35 to 51 years (adult group, n=4) were studied. Skin biopsies were obtained from these patients, who had developed HSc characteristized by raised, erythematous, pruritic, thickened, and noncompliant scars confined to the site of thermal injury that were treated at our hospital. Full-thickness skin samples were removed from these patients at the time of reconstructive surgical procedures, 1 to 2 years after burn injury. Normal, uninjured full-thickness skin from the adjacent tissue of each patient served as a control. Following the approval of the protocol by the Institutional Ethics Review Board, informed consent was obtained from the patients or their relatives. All procedures to obtain the tissues were approved by the Institutional Review Board at the 304th clinical department of the General Hospital of PLA.

To clearly identify the epidermal stem cells and the different portion of sweat glands in skin, we used two monoclonal antibodies, keratin 19 (K19, Clone: Ks 19.1, Maxin Biotec, Inc., USA) and keratin 14 (K14, Clone: LL002, Maxin Biotec, Inc., USA). They are thought to be the markers of human skin stem cells and the indicators of basal cells to the terminally differentiated epithelial cells of the cornified layer.[5] Also, K19 was used to identify the coiled gland secretory cells in human skin and K14 was used as the indicator of eccrine duct cells.[7,8] Streptavidin-peroxidase (SP) kits were purchased from the Company of Santa Cruz Biotech, USA.

Immunohistochemical staining
Indirect immunohistochemical staining with mouse monoclonal antibodies of K19 and K14 were performed on formalin-fixed, paraffin-embedded skin sections with the use of a staining system from Santa Cruz Biotech, USA. The sections were stained according to the manufacturer’s instructions. After rehydration, the skin sections were immersed in 0.3% H2O2 to block endogenous peroxidase activity, and extraneous antibodies labeling was blocked by 2% goat serum. The sections were then incubated with each primary antibody mentioned above in a moist chamber for 1 hour at room temperature, and washed 3 times with PBS. The relevant secondary antibodies were applied for 30 minutes, and then sections were thoroughly washed, incubated with the SP complex for 10 minutes, and developed according to the manufacturer’s instructions. After immuno-peroxidase staining, the sections were counterstained with hematoxylin for 1 minute at room temperature, rehydrated and mounted. As negative controls, slides were incubated with PBS instead of primary antibodies.

Histological evaluation
Excised wounds were immediately bisected and placed in 10% buffered formalin and then imbedded in paraffin. Sections (5 μm) were cut, mounted, and stained with hematoxylin and eosin.


Histological evaluation of sweat glands in normal and hypertrophic scar
In normal skin biopsies, hematoxylin and eosin staining showed that there was a clear skin structure including epidermis, dermis and subcutaneous tissues. Histologically, the normal mature epidermis consisted of five distinct layers, i. e. the stratum basal, stratum spinosum, stratum granulosum, stratum lucidum and stratum corneum. In the dermis and subscutaneous tissues, collagen fibers were organized normally and clearly demarcated. Most microvessels were visible. The hair follicle bulges and sweat gland structures were observed ( Fig. 1A ). It was found that the sweat gland consisted of a coiled gland in the lower portion of the dermis, lying at the border of the dermis and subcutaneous fat, connected to the surface by a duct that penetrates into the epidermis to permit expulsion of the gland’s secretory products ( Fig. 1A and B ). However, in hypertrophic scar, there was a rich vasculature, high mesenchymal density with thickened epidermal layer. Collagen fibers were organized in swirls with few macrophages, whereas some lymphocytes and eosinophiles were found ( Fig. 1C ). Hematoxylin and eosin staining revealed few coiled secretory glands under the portion of the scar tissue, which were expanded and organized in an irregular way and distributed under the border between the scar tissue and subcutaneous tissue ( Fig. 1D ). Nevertheless, it was hard to identify the eccrine ducts by hematoxylin and eosin staining of the scar tissue. There was no difference in sweat gland distribution in both young and adult groups.

Immunohistochemical evaluation of epidermal stem cells and sweat glands in normal skin and hypertrophic scar
Because the K19 is thought to be the stem cell marker for basal stem cells, hair follicle bulges stem cells and secretory cells in coiled glands, and K14 is the marker for eccrine duct cells, the two substances were used as the indicators for further identification of these cells in this study.[5,6] The Immunohistochemical results showed that strong K19 positive cells could be found in the secretory portion of sweat gland and some of the basal cells in normal skin, meaning that these K19 positive cells were stem cells or stem-like cells ( Fig. 2A and B ). Also, strong K14 positive cells were observed in some of the cuboidal epidermal cells in duct portion of the sweat gland, indicating that these K14 positive stain cells were stem cells or stem-like cells ( Fig. 2C ). In hypertrophic scar tissue, the different distribution of K19 and K14 positive stain cells could be observed. In the portion of the epidermis, the K19 positive cells were seen in some of the basal cells, however, these cells were scatterly distributed and just showed a weaker positive expression. In the portion under the scar tissue, a large number of coiled secretory glands were found. They were expanded and organized in an irregular way as described above. They were distributed under the border between the scar tissue and subcutaneous tissue ( Fig. 2D ). It was very interesting to find that although there was an increased thickness of epidermal layer and excessive accumulation of granulation tissue in scar tissue, there are still some tube-like K14 positive expression cells in the lower portion of the scar tissues. These K14 strong positive cells were seen in some of the cuboidal epidermal cells in duct part of the sweat glands, indicating that these K14 positive stain cells were stem cells or stem-like cells ( Fig. 2E and F ).


The objective of this study was to observe the patterns of sweat glands and epidermal stem cells in normal skin and scar tissue. The data of this study show that although there is no difference in distribution pattern of sweat glands in scar tissue between children and adults, a rich mass of secretory portion of sweat glands could be observed under the scar tissue. Also, some of the duct portion of the sweat glands were also visible in scar tissues. These results indicated that there are active sweat gland regeneration process during regeneration of granulation tissue and scar formation and a possibility for regeneration of sweat glands after extensive burn injury.

The questions are where do these sweat glands come from? Is there a possibility to regenerate a new sweat gland from adult epidermal stem cells or remained sweat glands in the wound bed? How to stimulate the regeneration of sweat gland? Usually, mammalian skin regenerates first, and repairs later. Fetal skin regenerates its original architecture after wound. In late stage of gestation, the skin changes its response to injury from regeneration to repairment by forming scar tissue as that of an adult. Scarless wound healing is often observed in the early stage of fetal development, while scar formation is common in adults.

As discussed above, epidermal stem cells are the resource of the epidermis and skin appendages such as hair follicles and sweat glands. The tissue regeneration by means of stem cells or progenitor cells depends on cell phenotype, as well as on their molecular environment, which plays a critical role.[3,9-11] In fetal wounds, the scarless wound healing own to a different molecular environment from that in adult. First, fetal wound does not exhibit an inflammatory phase, even though neutrophils and macrophages may infiltrate into the wound. Second, the normal pattern of collagen deposition and organization in fetal wounds is associated with three differences in the ECM from adult wounds, which diminish fibroblast proliferation and cross-linking of type Ⅰ collagen fibrils. Collagen and noncollagen protein synthesis are both elevated above the normal level, but there is no excessive deposition of type Ⅰ collagen fibrils and they exhibit the normal reticular organization. Third, epithelialization is more rapid than the migration of fibroblast into the wound.[9-11] In extensive burn injury, the skin with its appendages, such as hair follicles and sweat glands are damaged. The tissue repair cells including dermal fibroblast phenotype and wound healing molecular environment are quite different from those in fetal wounds. In normal dermal microenvironment, fibroblasts are heterogeneous with respect to their synthesis of ECM components as well as proliferation rate. Some are high producers of collagen. After trauma or an inflammatory response in the skin, wound healing exhibits an inflammatory phase, cells of the immune system are recruited and produce soluble factors which may stimulate the fibroblast subset producing large qualities of collagen and other ECM components.[12,13] The synthesis and degradation of collagen are essential to wound healing. These stimuli may cause the fibroblasts to proliferate or further upregulate their already high collagen production. The deposition of collagen results in the formation of a fine line scar, which restores much of the tensile strength to the injured tissue and is cosmetically acceptable. In certain individuals, however, the result of wound healing is the excessive accumulation of collagen, resulting in a hypertrophic scar. Histologically, the hypertrophic scar tissue differs from normal skin by its rich vasculature, high mesenchymal density, and thickened epidermal layer. Collagen fibers are organized in swirls. Usually, few macrophages are present, whereas some lymphocytes and eosinophils are found.[14] From the comparative results between fetal and adult wounds, we conclude that the changes in cell phenotype and microenvironment are the main obstruction for skin and its appendages regeneration.

We hypothesize that there is an active process of sweat gland regeneration during wound healing. Because of the changes of cell phenotype, molecular environment, as well as the increased thickness of epidermal layer and scar tissue, the regenerated sweat glands cannot establish an intact three-dimensional organization via the division and terminal differentiation of sweat gland cells. Also, few regenerated eccrine ducts cannot penetrate the increased epidermis and scar tissue to permit expulsion of the gland’s secretory products. This is why we can find rich mass of expanded secretory portion of sweat glands under the scar tissue and few duct portions in the scar tissue. Another possible question, whether these eccrine ducts are fragments of remained eccrine ducts in granulation tissues, need to be further clarified. According to our data, these eccrine ducts could be observed in series sections, and these eccrine ducts have the intact structure and positively stained with K14, meaning that these cells are alive and may regenerate from remained eccrine duct cells.

Finally, we speculate if there is a possibility to establish a three-dimensional organization via some active methods. We have found that the excessive accumulation of connective tissue in hypertrophic scar is an obstruction for sweat gland regeneration. Thus, it may be possible to stimulate the sweat gland regeneration by inhibiting excessive accumulation of granulation tissue and accelerating the division and terminal differentiation of the sweat glands via early eschar or cut off the excessive granulation tissue in the early wound healing stage, which may help to re-establish the good molecular environment and benefit the sweat gland regeneration from remained sweat glands in the wounded bed.[15,16] Another way is to induce the epidermal stem cells to directly differentiate into sweat gland cells. From the developmental biology, there are at least two most important fundamental links between the organogenesis and its regulation during embryogenesis, one is the action of matrix metalloproteinases regulated by a series of growth factors, another is the cellular-motion.[5,6] In this field, we are very interesting in the action of epidermal stem cells in skin. Usually, cutaneous epidermal stem cells play an important role in the skin regeneration. They are considered the key resource for epidermis and skin appendages, such as hair follicles and sweat glands. In normal epidermis, these stem cells are a single layer situated in the basement membrane of the epidermis or in the hair follicles and sweat glands. It is estimated that these cells comprise 1%-10% of the basal layer in vivo. According to their most salient features, namely slowly-cycling nature and high proliferative potential, in vivo and in vitro can identify them[5,6,17] Based on the relationship between stem cells and sweat gland cell development, we hypothesize that there is a possibility to induce the epidermal stem cells to directly differentiate into sweat gland cells and to establish a three-dimensional organization. Now, our groups are studying the morphology of epidermal stem cells and sweat gland development and have found some evidences to induce the sweat gland development from epidermal stem cells.[18] Furthermore, we are studying the interaction between epidermal growth factor and matrix metalloproteinases and their action in inducing the sweat gland development in human skin. The data from the primary study support the concept that the sweat gland can be re-established by inducing the stem cell division.[19-23]


1.Fu XB, Shen ZY, Guo ZR, et al. Healing of chronic cutaneous wounds by topical treatment with basic fibroblast growth factor. Chin Med J 2002;115:331-335.
2.Fu XB, Shen ZY, Chen YL, et al. Recombinant bovine basic fibroblast growth factor accelerates wound healing in patients with burns, donor sites and chronic dermal ulcers. Chin Med J 2000;113:367-371.
3.Stocum DL. Tissue restoration: approaches and prospects. Wound Rep Reg 1996;4:3-15.
4.Ladin DA, Garner WL, Smith DJ. Excessive scarring as a consequence of healing. Wound Rep Reg 1995;3:6-24.
5.Schon M, Bnwood J, O’Connell-Willstaedt T, et al. Human sweat gland myoepithelial cells express a unique set of cytokeratins and reveal the potential for alternative epithelial and mesenchymal differentiation states in culture. J Cell Sci 1999;112:1925-1936.
6.Matsuzaki T, Yoshizato K. Role of hair papilla cells on induction and regeneration processes of hair follicles. Wound Rep Reg 1998;6:524-530.
7.Turksen K, Troy T. Epidermal cell lineage. Biochem Cell Biol 1998;76:889-898.
8.Cotsarelis G, Kaur P, Dhouailly D, et al. Epithelial stem cells in the skin: definition, markers, localization and functions. Exp Dermatol 1999;8:80-88.
9.Streuli, C. Extracllular matrix remodeling and cellular differentiation. Curr Opin Cell Biol 1999;11:634-640.
10.Olutoye OO, Cohen IK. Fetal wound healing. Wound Rep Reg 1996;4:66-67.
11.Frantz FW, Diegelmann RF, Mast BA, et al. Biology of fetal wound healing: collagen biosynthesis during dermal repair. J Ped Surg 1992;27:945-949.
12.Lin RY, Sullivan KM, Argenta PA, et al. Scarless human fetal skin repair in intrinsic to the fetal fibroblast and occurs in the absence of an inflammatory responses. Wound Rep Reg 1994;2:297-305.
13.Wang R, Ghahary A, Shen Q, et al. Hypertrop hic scar tissue and fibroblasts produce more transforming growth factor-β1 mRNA and protein than normal skin and cells. Wound Rep Reg 2000;8:128-137.
14.Wang R, Ghahary A, Shen YJ, et al. Nitric oxide synthase expression and nitric oxide production are reduced in hypertrophic scar tissue and fibroblasts. J Invest Dermatol 1997;108:438-444.
15.Sempowski GD, Borrello MA, Blieden TM, et al. Fibroblast heterogeneity in the healing wound. Wound Rep Rep 1995;3:120-131.
16.Watt FM, Hogan BLM. Out of Eden: stem cell and their niches. Science 2000;287:1427-1430.
17.Jones PH. Epithelial stem cells. Bioessays 1997;19:683-690.
18.Fu XB, Sun XQ, Li XK, et al. Dedifferentiation of epidermal cells to stem cells in vivo. Lancet 2001;358:1067-1068.
19.Li JF, Fu XB, Sheng ZY. The interaction between epidermal growth factor (EGF) and matrix metalloprteinasas induces the development of sweat glands in human fetal skin. J Surg Res 2002;106:258-263.
20.Thomson JA, Itkovitz-Eldor J, Shapiro SS, et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145-1147.
21.Shamlott MJ, Axelwan J, Wang S. Derivation of pluripotent stem cell from ultured human primordial germ cell. Proc Natl Acad Sci U S A 1998;95:13726-13731.
22.Ferraris C, Chevalier G, Favier B, et al. Adult corneal epithelium basal cells possess the capacity to activate epidermal, pilosebaceous and sweat gland genetic programs in response to embryonic dermal stimuli. Development 2000;127:5487-5495.
23.Li A, Simmons PJ, Kaur P. Identification and isolation of candidate human keratinocyte stem cells based on cell surface phenotype. Proc Natl Acad Sci U S A 1998;95:3902-3907.

  1. National Basic Science and Development Programme (973 Programme),G1999054204;National Outstanding Young Researchers (No. 39525024), and the National Natural Science Foundation of China ().,No. 30170966, 30230370;