Lin Qiu, Xian-qin Jin, Dai-li Xiang, Yue-xian Fu, Xiao-fei Tian

Department of Burns and Plastic Surgery, Children’s Hospital, Chongqing Medical University, Chongqing, People’s Republic of China

SUMMARY. Objective: To investigate the difference in ratio of collagen in hypertrophic scar (HS) at different age periods and its causes, in order to provide a theoretical base for its special clinical treatment. Method: The expression of type I, type III collagen, and transforming growth factor-‚1 (TGF-‚1) was measured immunohistochemically in different-age HS and normal skin (NS). All the quantities were analysed. Messenger RNA (mRNA) expressions of collagenase (MMP-1) and tissue inhibitor of metalloproteinase (TIMP-1) were measured with the in situ hybridization technique. Results: 1. The ratio of type I/type III collagen and the quantification of type I collagen in HS increased compared with NS. The ratio of type I/type III collagen in the 1-19 years old HS group increased compared with the 20-50 years old group. 2. The expression of TGF-‚1 protein in HS was enhanced. The expression of TGF-‚1 in the 1-19 years old HS group significantly increased more than in the 20-50 years old group. 3. A significant increase of expression of TIMP-1 mRNA in HS was observed, but the expression of MMP-1 was lower both in HS and in NS. The expression of TIMP-1 mRNA and the expression of MMP-1 in the HS 1-19 years old group and the 20-50 years old group presented no difference. Conclusion: 1. The significant increase in TGF-‚1, which stimulates fibroblast to synthesize more type I collagen and enhances the ratio of type I/type III collagen, may be a cause of the higher incidence of scarring in the 1-19 years old group (children and teenagers). 2. The higher expression of TIMP-1 mRNA and the lower expression of MMP-1 mRNA are among the factors causing HS. Expressions of TIMP-1 and MMP-1 in HS have no direct relationship to age.

Introduction

Hypertrophic scar (HS) following burn injury or other wounds is a common, disfiguring, and functionally limiting form of dermal fibrosis. Clinically, it is characterized by excessive dermal fibrosis and scarring resulting from the imbalance between collagen synthesis and degradation during wound healing. The incidence of hypertrophic scar is related to age and race factors. The phenomenon may be caused by many factors of collagen synthesis and degradation. Collagen is the fundamental substance for the healing of wounds. As a kind of growth factor, transforming growth factor beta-1 (TGF-‚1) is very important in wound healing, but an over-expression of it may disturb the balance between synthesis and degradation of collagen and cause the deposition of extracellular matrix (ECM).1 TGF-‚1 induces the expression of ·-smooth muscle actin, stimulates the synthesis and maturity of type I collagen,2 and causes the formation of HS and scar contracture. Conversely, over-expression of TGF-‚1 may reduce collagen degradation by inhibiting the expression of collagenase (MMP-1) and increasing the production of the tissue inhibitor metalloproteinase (TIMP-1) and degradation products of collagenase. MMP-1 is the key enzyme in the degrading of type I and type III collagen in scars.3 A high expression of TIMP-1 and inhibition of MMP-1 activity decrease the degradation of collagen and cause the formation of HS. The pathogenesis is not well understood. In order to provide a theoretical basis for the special clinical treatment of scars, the difference in the ratio of collagen in hypertrophic scar of different ages and its causes was investigated.

Materials and methods

Tissue samples
Samples taken from thirty hypertrophic burn scars in eighteen male and twelve female patients were studied, as also normal control skin from the same patients. All were obtained during plastic or constructive surgery. The ages of the patients ranged from 1 to 50 years old. Hypertrophic scars were divided into two groups: 1-19 years old and 20-50 years old. Each group was divided into three stages: 6-12 months, 13-24 months, and over 24 months since burning. The scars were clinically symptomatic, characterized by itching, redness, and/or contraction, and the diagnoses were subsequently confirmed by biopsy. The patients had no history of systemic disease or systemic sclerosis, and they had not been treated with any hormones, radiation, or other therapy.

Antibodies and oligonucleotide probes
Rabbit polyclonal antibodies were used in order to recognize the protein of type I collagen and type III collagen, which was obtained from the Boshide company of Wuhan (China). The TGF-‚1 polyclonal antibodies came from the Santa Cruz company (USA). The following human sequence-specific oligonucleotide probes were used for hybridizations: for MMP-1, a 1.5-kb collagenase oligonucleotide; and for TIMP-1, a 0.6-kb metalloproteinase oligonucleotide. The sequences for these probes were respectively as follows: ¨ 5’-d(TTGTA GTAGT TTTCC AGGTA TTTCC AGACT) ¤ 5’- d(CATCA GGCAC CCCAC ACCTG GGCTT CTTCA) for collagenase; ¨ 5’-d(ACAGC AACAA CAGGA TGCCA GAAGC CAGGG) ¤ 5’-d(ACAAG CAATG AGTGCCACTC TGCAG TTGC) for metalloproteinase.

Immunohistochemistry
Immunohischemistry was performed using the SABC (StreptAvidin-Biotin) method. Paraffin human skin and scar sections were dewaxed with decreased concentrations of ethanol. Slides were immersed in a solution of 3% H2O2 containing distilled water in order to inactivate the endogenetic peroxidase for 10 min at room temperature. 20 Ìl of compound digestive juice was dripped onto sections for 20 min, and rinsed three times with distilled water. 20 Ìl of blocking of goat blood serum was pipetted for 20 min and then moved, after which primary antibody (anti-human antibody) was pipetted for 3 h at 20 °C. The sections were then rinsed with 0.1M phosphate-buffered saline (PBS). Secondary antibody (anti-animal body) was pipetted for 30 min at 20 °C and rinsed. The sections were immersed in SABC solution for 20 min at 20 °C and rinsed. Compounds of antigen-antibody were stained by immersing the samples in DAB solutions for 5 min; positive yellow grains were observed under a microscope and then counterstained with haematoxylin, dehydrated with increasing concentrations of ethanol, cleared in xylene, and mounted. Negative controls consisted of using fresh PBS instead of the primary antibodies.

In situ hybridization
Frozen human skin and scar sections, 20 Ìm thick, were fixed by fresh 4% paraformaldehyde and 1/1000 DEPC in 0.1M PBS for 40 min at room temperature, and rinsed three times. The slides were immersed in a solution of 0.5% H2O2 containing formaldehyde in order to inactivate the endogenetic peroxidase for 30 min, and then rinsed with distilled water. To expose mRNA nucleic acid, 40 Ìl of pepsin in 3% citromalic acid was added for 60 sec at 37 °C, and then rinsed with 0.5M PBS. The sections were prehybridized for 4 h at 40 °C with 20 Ìl of prehybridization buffer in humidified boxes containing 20 ml of 20% glycerine. The prehybridization solution was removed and 20 Ìl of hybridization buffer was pipetted onto each section and then covered with a cover slip. Hybridization was continued for 48 h at 40 °C in humidified boxes. Following hybridization, the sections were rinsed twice with 2¥ standard saline citrate (SSC) for 5 min and once with 0.5¥ and 0.2¥ SSC for 15 min at 37 °C respectively. After hybridization, the 40 Ìl of blocking solution, biotin anti-rat cardiox, SABC, and biotin peroxidase were pipetted in turn onto sections for about 30 min at 37 °C, and then rinsed with 0.5M PBS. The c-DNA-mRNA hybrids were stained by immersing the samples in DAB solutions for 15 min, observed under a microscope, counterstained with haematoxylin, dehydrated with increasing concentrations of ethanol, cleared in xylene, and mounted. The resulting brown grains were examined as positive express.

Morphometric analysis Tissue sections were studied with light microscopy using the computer colour image analysis system software. Five contiguous, non-overlapping microscopic fields of each section were analysed. For immunohistochemistry data, the quantity of type I collagen and type III collagen was measured with the ratio of positive and total area at a magnification of 10¥, and TGF-‚1 positive cells were counted in 200 total cells at a magnification of 100¥. For the in situ hybridization data, the results of MMP-1 and TIMP-1 were analysed in the same way as TGF-‚1.

Statistical analysis All data were analysed using SAS6.12 statistics software, analysis of variance, Q-test, and t-test.

Results

Observation of type I collagen and type III collagen expression in hypertrophic scar and normal skin tissue and data analysis (Table I) <% createTable "Table I ","Changes of type I and III collagen in HS and control NS groups (X– ± SD, %)",";Collagen§1,3§1-19 yr old; §1,3§20-50 yr old; @;Type§1,3§HS (months after thermal injury);NS§1,3§ HS (months after thermal injury);NS@; 6-12 m;12-24 m;> 24 m; 6-12 m;12-24 m;> 24m ; @; I;53.6 ± 3.26*;55.3 ± 2.46*;49.4 ± 3.80*;28.6 ± 2.51;46.4 ± 2.31*;48.6 ± 2.50*;45.0 ± 2.39*; 29.5 ± 1.97*@;III;8.12 ± 0.82*; 8.85 ± 1.21*;7.48 ± 0.82*;22.7 ± 1.56*;12.5 ± 2.04*; 13.3 ± 1.73*; 11.5 ± 1.69*; 21.2 ± 1.34@; I/III;6.60;6.25;6.60;1.26;3.71;3.65;3.91;1.39@§1,9§* p < 0.05","",4,300,true %> The mass of yellow positive signal of type I collagen was located in intercellular substance in the dermis of hypertrophic scar, and a small quantity of one was observed in fibroblasts and blood vessels. In normal skin, the distribution of type I collagen was approximately the same as in hypertrophic scar, but there were positive signals in fibroblasts and blood vessels. Scattered, thin, and yellow type III collagens in intercellular substance, fibroblasts, and blood vessels in the dermis of hypertrophic scar were observed. Except for more intensive type III collagen, those in normal skin were similar to hypertrophic scar.

In all hypertrophic scar tissue, more brown positive expressions for TGF-‚1 mRNA were found. The majority were in fibroblasts, and the minority were in endothelial cells. The most positive cells were distributed in nodular of collagen. The thick positive grains were located in cell plasm and scattered in the intercellular substance. The positive cells for protein of TGF-‚1 were few in normal skin tissue.

Localization of MMP-1 and TIMP-1 mRNA and data analysis (Table III)

The expression of MMP-1 mRNA in hypertrophic scar and normal skin tissue was inferior, and a small quantity of yellow positive grains was located in nodular of collagen. <% createTable "Table III ","Changes of TGF-‚1 in HS and NS groups (X– ± SD, %)","; §1,2§1-19 yr old§1,2§20-50 yr old@; HS;NS;HS;NS@;MMP-1;0.43 ± 0.02;0.41 ± 0.02;0.45 ± 0.04;0.40 ± 0.02@;TIMP-1;3.55 ± 0.56*;0.32 ± 0.05;3.30 ± 0.40*;0.37 ± 0.04@;TIMP-1/MMP-1;8.26; 7.33; @§1,5§ *: p < 0.01","",4,300,true %> In fibroblasts, nodular of collagen, and endothelial cells of hypertrophic scar, the brown positive grains of expression of TIMP-1 mRNA were increased. The positive grains were scattered in endothelial cells in dermis.

Discussion

Collagen is the main component of ECM, and type I and type III collagen fibres are the capital parts in hypertrophic scar and normal human skin. Morphological observation reveals the negative correlation relationship between the diameters of collagen fibre.4 During the formation of HS, the increase of type I and decrease of type III are important, and the thick type I collagen is the fibrosis pathology background of scar tissue. The results of the experiment show that type I increased and type III decreased more obviously in the HS group than in the NS group, while their content and ratio in different periods (6-12 months, 12-24 months, and > 24 months) in the groups were same. In the 1-19 years old group (children and teenagers), type I increased more obviously in the HS group, and the ratio of type I/III was higher than in the 20-50 years old group. With the increase of type I collagen during the healing process and the decrease of type III collagen as argyrophilic fibre disequilibrium in the ratio of type I/III, the disequilibrium in the end causes hypertrophic scars. The imbalances in the ratio were more obvious in 1-19 years old group. The difference of the I/III collagen ratio in different age groups might be related to the change in factors regarding the synthesis and degradation of collagen. The results suggested that the imbalance in the collagen ratio in HS tissue was closely related to age, which implies that measures to decrease the content of type I collagen in tissue and control the ratio’s disequilibrium might reduce the formation of hypertrophic scars in teenagers. TGF-‚1 is one of the regulating factors causing ECM to be deposited in fibrosis diseases, and over-expression and local application of TGF-‚1 will cause local fibrous tissue proliferation.5 TGF-‚1 can accelerate the synthesis of collagen and stimulate the synthesis of fibre accretion protein, tendon protein, and proteoglycan by increasing the expression of type I and type III collagen genes. The previous study showed that the stable mRNA level of TGF-‚1 in HS tissue was apparently higher than in NS tissue. This suggested that HS tissue could itself synthesize TGF-‚1; also, TGF-‚1 stimulated fibroblast proliferation and promoted synthesis and deposition of type I collagen.6 Immunohistochemical results showed that the expression of TGF-‚1 protein in HS tissue was much higher than in NS tissue. It was highest in fibroblasts and vascular endothelial cells, and it was also expressed to a small extent in ECM. All this indicated that the expression of TGF-‚1 protein was very important for controlling the formation of HS. Also, in this study, the amount of TGF-‚1 was higher in the 1-19 years old group than in the 20-50 years old group, which indicated that the expression of TGF-‚1 in HS was closely related to age. The high incidence of HS in the 1-19 years old segment may be related to the high expression of TGF-‚1 during the wound-healing process. This was consistent with the result that the ratio of type I/III increased in this age segment. It has been shown that use of the neutralization antibody of TGF-‚1 can reduce the formation of scars during the wound-healing process and that the healed wound had the same corium constitution and tissue tension.7 The controlling mechanism that maintains the normal ECM constituent is very complicated - it includes keeping the balance between synthesis and degradation of collagen. MMP-1 and TIMP-1 are important enzymes related to collagen degradation. By binding with MMP-1 to form the complex, TIMP-1 inhibits the activity of collagenase and decreases collagen degradation. Some specialists3 have observed that the expression of MMP-1 mRNA and its enzyme activity in HS fibroblast decreased compared with that of NS fibroblast, and confirmed that the low expression of MMP-1 and the high expression of TIMP-1 in HS reduced collagen degradation and caused the over-collecting of collagen fibre. In the experiment, the hybridization in situ showed that the expression of TIMP-1 mRNA was much higher in HS tissue than in NS and that the expression of MMP-1 was much lower. This suggests that low expression of MMP-1 and high expression of TIMP-1 could inhibit collagen degradation and induce scar formation. The expressions of TIMP-1 mRNA and MMP-1 mRNA did not differ between the 1-19 years old HS group and the 20-50 years old HS group, which means that the expression of collagen degradation factors in HS may not be directly related to age.

Conclusion

It may therefore be concluded that TGF-‚1 could facilitate synthesis of type I collagen and the imbalance in the ratio of I/III collagen, and that the high expression of TIMP-1 could inhibit the activity of MMP-1 and reduce collagen degradation. The imbalance between the synthesis and the degradation of collagen would lead to HS formation. In the 1-19 years old HS group, the high ratio of I/III collagen might be caused by the high expression of TGF-‚1, which increased the synthesis of type I collagen; however, the mechanism is not clear and awaits further study. Further research is necessary to find the key factors linked to scar formation and to provide theoretic evidence for the control, prevention, and cure of scars.

RESUME. But: Investiguer la différence du rapport de collagène dans les cicatrices hypertrophiques (CH) de divers âges et ses causes, afin d’offrir une base théorique pour le traitement spécial de ces cicatrices. Méthode: L’expression du collagène de type 1, de type III et du facteur de croissance TGF-‚1 a été mesurée moyennant l’immunohistochimie dans des CH à divers âges et dans la peau normale (PN). Toutes les quantités ont été analysées. Les expressions du RNA messager (mRNA) de la collagénase (MMP-1) et de l’inhibiteur tissutal de la metalloprotéinase (TIMP-1) ont été mesurées moyennant la technique de l’hybridisation in situ. Résultats: 1. Le rapport du type I/type III collagène et la quantification du collagène de type I dans les CH augmentaient en comparaison avec la PN. Le rapport de collagène type I/type III dans le groupe CH âgé 1-19 ans augmentait par rapport au groupe âgé 20-50 ans. 2. L’expression de la protéine TGF-‚1 dans les CH augmentait. L’expression de TGF-‚1 dans le groupe CH âgé 1-19 ans augmentait en manière significative plus que le groupe âgé 20-50 ans. 3. Une augmentation significative a été observée dans l’expression de TIMP-1 mRNA dans les CH, mais l’expression de MMP-1 était inférieure soit dans les CH soit dans la PN. L’expression de TIMP-1 mRNA et l’expression de MMP-1 dans le groupe CH âgé 1-19 ans et dans le groupe âgé 20-50 ans ne présentaient aucune différence. Conclusion: 1. L’augmentation significative du TGF-‚1, qui stimule le fibroblaste à synthétiser une quantité majeure de collagène de type I et augmente le rapport de collagène type I/type III, peut être une des causes de l’incidence plus élevée de la cicatrisation dans le groupe âgé 1-19 ans (enfants et adolescents). 2. L’expression plus élevée de TIMP-1 mRNA et l’expression moins élevée de MMP-1 mRNA constituent des facteurs qui contribuent à la création des CH. Les expressions de TIMP-1 et de MMP-1 dans les CH n’ont aucun rapport direct avec l’âge.

Bibliography

1. Richard Y., Lin B.S.: Exogenous transforming growth factor ‚ amplifies its own expression and induces scar formation in model of human foetal skin repair. Ann. Surg., 222: 146-54, 1995.
2. Grande J.P., Melder D.C.: Modulation of collagen gene expression by cytokines: Stimulatory effect of transforming growth factor ‚1, with divergent effects of epidermal growth factor and tumour necrosis factor-alpha on collagen type I and collagen type IV. J. Lab. Clin. Med., 130: 476-86, 1997.
3. Ghahary A., Shen Y.J.: Collagenase production is lower in post-burn hypertrophic scar fibroblasts and is reduced by insulin-like growth factor-1. J. Invest. Dermatol., 106: 476-81, 1996.
4. Linares H.A.: From wound to scar. Burns, 22: 339-52, 1996.
5. Rong H., Tang X.M.: Post-surgical intraperitoneal exposure to glove powders modulates inflammatory and immune-related cytokine production. Wound Rep. Reg., 5: 89-96, 1997.
6. Zhou L.J., Ono I.: Role of transforming growth factor ‚1 in fibroblasts derived from normal and hypertrophic scarred skin. Arch. Dermatol. Res., 298: 646-52, 1997.
7. Shah H.: Control of scarring in adult wound by neutralizing antibodies to TGF-‚. Lancet, 1: 213-4, 1992.