Annals of Burns and Fire Disasters - vol. Pendig Publication - n. 2 - June 2005

ULTRASTRUCTURAL DIFFERENTIATION OF ABNORMAL SCARS

Meenakshi J.¹, Jayaraman V.², Ramakrishnan K.M.³, Babu M.¹

¹ Biomaterials Division, Central Leather Research Institute, Adyar, Chennai, India
² Division of Plastic Surgery, Kilpauk Medical College and Hospital, Chennai
³ Plastic Surgery and Burn Intensive Care Unit, K.K. Childs Trust Hospital, Chennai

SUMMARY. Aim: To evaluate the differences between keloid and hypertrophic scars by biochemical and ultrastructural techniques. Method: Over 1000 patients with different types of scars were studied and followed up for a period of 20 years. The histochemical and biochemical analysis with respect to the composition of the extracellular matrix of the dermis was conducted. At the ultrastructural level, collagen deposition and assembly were studied using electron microscopy. The rate of proliferation and metabolic activity of the dermal fibroblasts isolated from the normal skin and scar biopsies were studied to assess the cause of excess matrix deposition in scar tissues. Results: Evaluation of different types of scars showed that both keloid and hypertrophic scars have excess matrix deposition in terms of collagen and proteoglycans. Keloid shows a high amount of acid-soluble collagen. The assembly of collagen fibrils is also abnormal in keloids. Studies on the proliferation and metabolic activity showed that keloid fibroblasts have a higher rate of proliferation and metabolic activity than fibroblasts from hypertrophic scars and normal skin. Finally, keloid fibroblasts show high and intense staining for the endoplasmic reticulum, suggesting a possible reason for high activity of these fibroblasts. Conclusion: Keloids and hypertrophic scars show distinct ultrastructural patterns of both collagen deposition and assembly. These parameters could be refined by further research, and they would thus serve as a useful tool for surgeons to distinguish different types of scars and adopt suitable therapeutic strategies.

Introduction

The process of wound healing is initiated in the skin when there is a disturbance in the form of surgical injury or trauma of any sort, such as a burn. Wound healing is a process involving a delicate equilibrium between synthesis and degradation of extracellular matrix (ECM) components.1 In certain individuals, owing to certain racial and genetic factors, matrix deposition during wound healing is exuberant, resulting in scars that are unsightly and problematic for treatment. These scars are usually either hypertrophic or keloid, or they belong to an indeterminate group.

Keloids and hypertrophic scars are fibroproliferative disorders resulting from abnormal wound healing. Clinically, keloids are defined as scars that extend beyond the margins of the original wound, while hypertrophic scars remain confined. Keloids can occur spontaneously in individuals with a predisposition to keloid formation. They recur after surgical excision or with skin grafting and do not regress with time. Hypertrophic scars, on the other hand, can be corrected by surgical procedures and they regress or flatten over a period of time.3 Apart from these well-defined forms of scars, there is an indeterminate group of scars with a massive, raised, keloid-like appearance that do not recur after surgical excision.2 Thus, clinical evaluation based merely on appearance is not very helpful for the differentiation of aberrant scars.

Although the process of wound healing has been well studied, the underlying mechanism leading to the formation of abnormal scars remains obscure. However, it is known that any injury to the reticular layer of the dermis leads to the formation of such scars, and it is in this region that fibroblasts reside. Hence, dermal fibroblasts are considered to be key players in scar formation.

Apart from fibroblasts, ECM components also parti-cipate in scar formation. Scar tissue is characterized by overabundant ECM deposition in the dermal region of the skin and by an abnormal response of fibroblasts to growth modulators.5 However, in spite of these known aspects of abnormal wound healing, it is difficult for clinicians to differentiate such scars. A study based on the ultrastructural examination of ECM in the dermal region was therefore undertaken to examine the basic differences between the dermis of hypertrophic scars, keloids, and normal skin. The ECM of scar biopsies and normal skin biopsies was analysed using biochemical, histochemical, and ultrastructural methods. The results of this study could contribute to the armamentarium of physicians in the prevention of the formation of such scars and help to develop a treatment protocol.

Materials and methods

Clinical specimens

Tissues were obtained from the cosmetic surgery department of K.K. Childs Trust Hospital and Kilpauk Medical College and Hospital. Informed consent was obtained from each patient. All the patients providing tissue for this study were of Indian origin. The lesions were diagnosed as keloids or hypertrophic scars on the basis of clinical appearance and the history of the lesions.

Water content

The tissue specimens were immediately weighed for wet weight determination and lyophilized to obtain a constant dry weight. The difference between wet and dry weights was used as an estimate of water content.

Estimation of collagen

Estimation of hydroxyproline was used as a measure of collagen content in the tissues. Samples were hydrolysed in 6N HCl in sealed tubes at 110 ， for 22 h. The hydrolysed samples were evaporated to dryness in a boiling water bath to remove acid, and the residue was dissolved in distilled water and made up to a known volume. Hydroxyproline was determined by the method of Woessner.

Estimation of proteoglycans

Proteoglycans are made up protein cores that are heavily glycosylated with glycosamino glycans (GAGs). GAG estimation was therefore used as a measure of proteoglycan content in the tissues. GAGs were estimated by their hexosamine content. Samples were hydrolysed in 2N HCl in sealed tubes at 110 ， for 6 h. The hydrolysed samples were evaporated to dryness in a boiling water bath to remove acid, and the residue was dissolved in distilled water and made up to a known volume. The total hexosamine content was estimated by the method of Elson and Morgan.

Preparation of tissues for electron microscopy

The tissues and cell pellets (fibroblasts isolated from biopsies) were fixed with 2.5% glutaraldehyde in 0.1M phosphate buffer, pH 7.2, at room temperature. Post-fixation in 1% osmium tetra oxide in phosphate buffer was carried out for a further period of 3 h at room temperature. Tissues fixed in osmium tetra oxide were then washed, dehydrated through ascending grades of acetone, and embedded in araldite in flat embedding models, appropriate care being taken in orienting the tissues. Ultrathin sections were cut on the LKB ultratome V fitted with a glass knife and collected on copper grids. These grids were stained with 2% uranyl acetate and post-stained with 2% lead ci-trate solution. Stained grids were examined using a JEOL 100CX electron microscope.

Isolation of collagen

Skin and scar biopsies from human subjects were used in our study. All specimens were frozen at -20 ， shortly after removal. Prior to extraction, the material was coarsely minced and thoroughly washed in 0.5M sodium acetate at 4 ，, sedimented, and resuspended in 0.5M acetic acid for three days to extract collagen. The supernatant was precipitated with 5% NaCl and centrifuged. The pellet was redissolved in 0.5M acetic acid and dialysed against 0.02M disodium phosphate. The precipitate formed was redissolved in 0.5M acetic acid and the collagen was purified by repeating the above process and finally lyophilized.9 The residue left after acid extraction was subjected to pepsin digestion (10-15 mg per g dry tissue weight). After 20 h, the suspension was centrifuged (70,000 g) for 1 h. The pellet was resuspended and digested again with pepsin twice. The supernatants were combined and subjected to sequential neutral salt precipitation.10 The pellet obtained with 0.7M NaCl was dissolved in 0.5 M acetic acid and dialysed against 0.05 M acetic acid and lyophilized. The salt-free lyophilized collagen was stored in a desiccator at 4 ，.

Segment-long-spacing (SLS) formation and electron microscopy

Purified type I collagen was dissolved in 0.5M acetic acid, and the solutions were centrifuged at 100,000 g for 3 h to remove insoluble materials. To form SLS crystallites, the solutions were dialysed against 0.2% ATP in 0.1M acetic acid at 4，. The aggregates were taken on collodion-coated grids, positively stained with 2% phosphotungstic acid, and examined under JEOL 1200EX-II electron microscope operating at 80 kV.

Isolation and culture of dermal fibroblasts and meas-urement of growth and metabolic rate

The portions of biopsied tissues collected in sterile culture media were washed thoroughly in sterile PBS (0.01M, pH 7.2) and treated with 0.5% trypsin 20 mM EDTA solution for 16 h at 4 ，. After removal of the separated epidermis, the dermis was finely chopped and treated with 200 units/ml of collagenase in the presence of DMEM and 10% foetal bovine serum (FBS) at 37 ， for 16 h. The digested tissue was centrifuged at 2000 rpm for 5 min and the pellet-containing cells were plated in a tissue culture flask containing a sufficient quantity of DMEM and 10% FBS. The plated cells were allowed to adhere and grow to confluence and were subcultured. In order to measure the growth rate, the fibroblasts were labelled, and 3H-thymidine was monitored at different time points. For the meas-urement of their metabolic activity, total protein content was measured at different time points.

Results

Cases studied

Scar samples were collected from patients admitted to the burns ward and the plastic surgery ward of two well-known hospitals in Chennai, India, as described in the materials and methods section. The scar samples were collected and the patients were followed for a period of 20 years. During this period, 1223 patients with keloids and 1338 patients with hypertrophic scars were studied (Table I).

Some of these scars were unsightly and led to difficulty or incapacity in the use of the affected organ (Fig. 1).

Biochemical composition of ECM of scars

The primary ECM components, namely water content, collagen, and proteoglycans, were analysed. Water content was estimated as the difference between the wet weight and dry weight of biopsied tissues. Collagen was estimated by the hydroxyproline content. Since GAGs are the major components of proteoglycans, the proteoglycan content of the biopsies was indirectly estimated by quantitation of GAGs. Scars showed higher amounts of all three components studied than normal skin. Among the scars, keloids showed a higher amount of these components (Table II).

Thicker dermis in keloids with abnormal deposition of collagen

The histological assessment of scar and skin biopsies was performed by routine haematoxylin- and eosin-staining. Both keloid and hypertrophic scars showed a marked absence of epidermal ridges. A very thick epidermis was observed in hypertrophic scars, while in keloids a thick dermis was observed. Both types of scars showed heavy infiltration of cells in the dermal region (Fig. 2).

More cells in the dermal region could contribute to greater deposition of ECM. The tissue sections were therefore subjected to electron microscopy in order to study collagen deposition in the ECM. Higher amounts of collagen deposition were observed in the ECM of both types of scars (Fig. 3).

However, in keloid cases, collagen fibrillation appeared to be abnormal (Fig. 3a).

Higher amounts of acid-soluble collagen in keloid

The extractability of collagen in different types of buffer is used as a measure for identifying the extent of cross-linking in a given collagen sample. Collagen samples isolated from the skin and scar biopsies were extracted with neutral salt solution and 0.5 M acetic acid and by pepsin treatment. Normal skin biopsies showed higher amounts of pepsin-soluble collagen, while keloids showed higher amounts of acid-soluble collagen. Hypertrophic scars showed only slight variations in the amounts of acid- and pepsin-soluble collagen (Table III).

A lower amount of pepsin-soluble collagen in keloids was an indication of lower levels of cross-linking in these samples. To confirm this, pepsin-soluble collagen from normal skin and keloids was subjected to SLS analysis by electron microscopy. This study clearly indicated that keloid collagen was less cross-linked than normal skin collagen (Fig. 4).

Keloid fibroblasts show more proliferation

Excess matrix deposition in cases of keloids and hypertrophic scars could be the result of a higher proliferation rate of dermal fibroblasts, of higher metabolic activity, or of a combination of both these events. Dermal fibroblasts were therefore isolated and propagated from the scar and skin biopsies. Keloid fibroblasts showed a higher proliferation than hypertrophic scars and normal skin fibroblasts (Graph 1).

The rates of DNA synthesis and protein synthesis were studied in order to analyse the metabolic activity of these fibroblasts. Both these studies showed that keloid fibroblasts were metabolically more active than fibroblasts from hypertrophic scars and normal skin (Table IV).

To test the hypothesis that a higher turnover of total proteins would lead to a higher turnover of secretory proteins that would eventually result in more matrix deposition, we studied the cytoplasmic architecture using electron microscopy. An increased and intense staining for the endoplasmic reticulum (ER) was obtained in keloid fibroblasts (Fig. 5), thus confirming that keloid fibroblasts are indeed metabolically more active.

Discussion

This study was undertaken to help clinicians differentiate between the different types of abnormal scars that develop as the final result of trauma and surgical injuries. Clinical investigations were conducted over a period of twenty years in two major hospitals in the city of Chennai, India. Over a thousand samples of each scar type were studied during this period, and the patients� progress was carefully followed and documented. The initial categorization of the scars was done by routine clinical procedures.

During such categorizations the medical community is posed with serious difficulties in the differentiation of the scars, and a comprehensive biochemical and ultrastructural analysis was therefore effected. The biopsied and operated scar specimens were first subjected to biochemical analysis of the major ECM components of the dermal region, namely collagen, water, and proteoglycans (Table II). All these components were found to be higher in keloids than hypertrophic scars and normal skin. The high water-content of the scar samples was indicated by magnetic resonance imaging of the patients (data not shown).

Patient follow-up revealed that the higher the water content of the scars, the longer it took them to regress. Excess water content could thus be one the causes for non-regression of keloids. Higher water content also indicates better extractability of ECM components. However, the easy extractability of ECM components also indicates poor matrix assembly. Excess water probably interferes with the fibrillar assembly of collagen. In normal skin the network-like assembly of collagen is maintained by the proteoglycans and GAGs. GAGs are responsible for the lateral assembly of collagen. Even though there is an excess amount of GAGs in scar tissues, the presence of excess water probably has an inhibitory effect on the lateral assembly of collagen, which is clearly seen in keloids (Fig. 3). Depending upon the association and assembly of collagen fibrils, collagen can be extracted in different types of buffers.12 This property of collagen was here exploited in order to study the extent of cross-linking (lateral association) of collagen fibrils in scar tissues.

Keloid collagen shows maximum extractability in acid (Table III), which indicates the lower extent of cross-linking between collagen fibrils. This cross-linking pattern was confirmed by the SLS analysis of pepsin-soluble collagen obtained from keloid and normal skin (Fig. 4). Our study therefore shows a distinct anomaly with respect not only to matrix deposition but also to matrix assembly in keloids.

The scar tissues showed a large number of cells in the dermal region (Fig. 2). The presence of more fibroblasts could lead to greater deposition of ECM. The high rate of matrix deposition could also result either from the high rate of proliferation of dermal fibroblasts or from metabolically more active fibroblasts. After studying these aspects, it was observed that both are high in keloid fibroblasts (Graph 1 and Table IV). The high rate of proliferation and metabolic activity could be the cause of excess matrix deposition in scar tissues. Transmission electron microscopic analysis of the dermal fibroblasts showed that keloid fibroblasts presented increased and intense staining for ER. The presence of more ER indicated that there was a high protein turnover, leading to more secretion of secretory proteins. There was therefore more deposition of ECM components (Fig. 5).

Conclusion

On the basis of this study it may be concluded that keloids and hypertrophic scars show distinct differences at the ultrastructural level. Excess matrix deposition, along with high water content, could lead to the abnormal assembly of collagen in keloids. Excess matrix deposition results from the high rate of proliferation and metabolic activity of dermal fibroblasts in keloids. Further research to probe the molecular mechanism underlying scar formation is necessary in order to design appropriate tools for the early detection of keloids and hypertrophic scars. Using these, it will be possible to develop stepwise strategies for suitable therapeutic interventions.

RESUME. But : Les Auteurs se sont propos駸 d帝valuer les diff駻ences entre les cicatrices ch駘o�des et hypertrophiques utilisant des techniques biochimiques et ultrastructurales. M騁hode : Plus de 1000 patients atteints de divers types de cicatrices ont 騁� 騁udi駸 et suivis pendant une p駻iode de 20 ans. Une analyse histochimique et biochimique a 騁� effectu馥 pour ce qui concerne la composition de la matrice extracellulaire du derme. Au niveau ultrastructural, la d駱osition et l誕ssemblage du collag鈩e ont 騁� 騁udi駸 en utilisant la microscopie 駘ectronique. Le taux de prolif駻ation et l誕ctivit� m騁abolique des fibroblastes dermiques isol駸 de la peau normale et des biopsies cicatricielles ont 騁� 騁udi駸 pour 騅aluer la cause de l弾xc鑚 de d駱osition de matrice dans les tissus cicatriciels. R駸ultats : L帝valuation des diff駻ents types de cicatrices a r騅駘� que soit les ch駘o�des soit les cicatrices hypertrophiques pr駸entent un exc鑚 de d駱osition matricielle pour ce qui concerne le collag鈩e et les prot駮glycans. La ch駘o�de poss鐡e une haute quantit� de collag鈩e acide-soluble. En outre, l誕ssemblage des fibrilles de collag鈩e est anormal aussi dans les ch駘o�des. Les 騁udes sur la prolif駻ation et l誕ctivit� m騁abolique ont d駑ontr� que les fibroblastes des ch駘o�des pr駸entent un taux de prolif駻ation et d誕ctivit� m騁abolique plus 駘ev� par rapport aux fibroblastes des cicatrices hypertrophiques et de la peau normale. Enfin, les fibroblastes des ch駘o�des pr駸entent une coloration 駘ev馥 et intense pour ce qui concerne le r騁iculum endoplasmique, ce qui pourrait sugg駻er une justification plausible de l誕ctivit� 駘ev馥 de ce type de fibroblaste. Conclusion : Les ch駘o�des et les cicatrices hypertrophiques pr駸entent des mod鑞es ultrastructuraux tr鑚 clairs soit de la d駱osition du collag鈩e soit de son assemblage. Ces param鑼res pourraient 黎re raffin駸 avec des recherches ult駻ieures et ils pourraient donc constituer un instrument utile pour les chirurgiens qui doivent distinguer les diff駻ents types de cicatrices et adopter des strat馮ies th駻apeutiques appropri馥s

Bibliography

1. Clark R.A.F.: 典he Molecular and Cellular Biology of Wound Repair�, Plenum Press, New York, London, 1996.
2. Tredget E.E., Nedelec B., Scott P.G., Ghahary A.: Hypertrophic scars, keloids and contractures - the cellular and molecular basis for therapy. Surg. Clin. Nor. Am., 77: 701-30, 1997.
3. Darzi M.A., Chowdri N.A., Kaul S.K. et al.: Evaluation of various methods for treating keloids and hypertrophic scars: A ten-year follow-up study. Br. J. Plast. Surg., 45: 374-79, 1992.
4. Diegelmann R.F.: Cellular and biochemical aspects of normal and abnormal wound healing: An overview. J. Urol., 157: 298-302, 1997.
5. Raghow R.: The role of extracellular matrix in post-inflammatory wound healing and fibrosis. FASEB J., 8: 23-831, 1994.
6. Woessner J.F., jr: The determination of hydroxyproline in tissue and protein samples containing small portions of this amino acid. Arch. Biochem. Biophys., 93: 440-7, 1961.
7. Elson L.A., Morgan W.T.J.: Colorimetric method for the determination of glucosamine and chondrosamine. Biochem. J., 27: 1824-8, 1933.
8. Gupta P.D., Vijayasaradhi S., Reddy A.G.: Keratinization of rat vaginal epithelium. III. Effect of estradiol on keratinization. Biol. Cell., 65: 281-9, 1989.
9. Chandrakasan G. Torchia D.A., Piez K.A.: Preparation of intact monomeric collagen from rat tail tendon and skin and the structure of the non-helical ends in solution. J. Biol. Chem., 251: 6062-7, 1976.
10. Miller E.J., Rhodes R.K.: Preparation and characterization of the different types of collagen. Methods Enzymol., 82: 33-64, 1982.
11. Kobayashi K., Ito T., Hoshino T.: Correlation between negative staining pattern and hydrophobic residues of collagen. J. Electron. Micros., 35: 272-5, 1986.
12. Babu M., Padma Bai, Suguna L., Ramachandran K., Ramakrishnan K. M.: Differentiation of keloid and hypertrophic scar: Correlation of water proton relaxation times with the duration of the scar. Physiol. Chem. Phys. Med. NMR, 25: 113-20, 1993.