Triterpenes from Centella asiatica stimulate extracellular matrix accumulation in rat experimental wounds

ARTICLE

Wound healing is a complex and ordered sequence of events which involves the activation of a large number of different cell types. Inflammation and angiogenesis, formation of granulation tissue, re-epithelialization, deposition of newly synthetized connective tissue macromolecules, and matrix remodeling are some of the main events involved in normal wound repair. A defect in any of these components of the repair process may lead to impaired wound healing [for review, see reference 1].

Centella asiatica is a plant which grows spontaneously around the Indian Ocean. It has been used for several centuries in the traditional medicine of India and oriental countries as a treatment for wounds. A drug derived from the plant has been developed in the European pharmacopea, under the name of "Titrated Extract from Centella asiatica" (TECA). It is a reconstituted mixture of 3 triterpenes purified from the plant, asiatic acid [2], madecassic acid [3] and asiaticoside [4].

Previous in vitro studies from our laboratory [5] and others [6, 7] demonstrated that TECA was able to stimulate collagen synthesis in fibroblast cultures, asiatic acid and asiaticoside being the most active among the 3 triterpenes. In vivo, it was demonstrated that asiaticoside and some asiaticoside mimetics were able to increase the tensile strength of incisional wounds in rats [8, 9].

In the present study, we used the wound chamber model described by Schilling et al. [10] to study the biochemical parameters of wound healing in vivo. This model creates a wound with a stable dead space whose content may be easily collected for analysis. We demonstrated that TECA and its invidual components may stimulate extracellular matrix macromolecule accumulation in the wound chamber and accelerate the healing process.

Methods

Animals

Male Sprague-Dawley rats weighing approximately 200 g were used in the experiments. They were provided by the Centre d'Élevage Dépré (Saint Doulchard, France). They were placed in individual suspended stainless steel cages with food and water ad libitum. All animals received humane care, in compliance with our institution's guidelines for the use of laboratory animals.

Reagents

Usual reagents (analytical grade) were from Prolabo (Paris, France). Hydroxyproline was from France Biochem (Meudon, France) and uronic acid from Sigma (La Verpillière, France). TECA and its separate components, asiatic acid, madecassic acid and asiaticoside were provided by Laboratoires Roche-Nicholas, division Serdex (Saint-Ouen, France). For injection into the wound chambers, they were solubilized in a mixture of polypropylene glycol/benzyl alcohol (10/1, v/v).

Wound chambers

Wound chambers were made of stainless steel wire mesh (C-CX20, EDMED Inc., Bellevue, WA), as 1 cm-diameter by 2.5 cm-long cylinders. They were closed at both ends by Teflon caps and sterilized by autoclaving.

Surgical procedures

Rats were anesthesized by intraperitoneal injection of sodium pentobarbital (40 mg/kg ; Clin-Midy, Paris, France). Dorsal hair was clipped in a wide band from the scapula to the pelvis and the nude area was sterilized with polyvidone iodine (Betadine^®; Laboratoires Sarget, Merignac, France). One incision was made perpendicular to the spine to the skin's full thickness, through the panniculus carnosus to the fascial plane. A space approximately the size of the chamber was opened under the dermis and sterile wound chambers with caps were slipped beneath the skin. The incisions were closed with individual 4.0 nylon sutures. Animals were then returned to their cages.

Experimental design

A first set of experiments was used for studying the effects of increasing amounts of TECA injected into the wound chambers. For that purpose, 28 rats were divided into 7 groups of 4 and one wound chamber was implanted on the back of each. The different series received the injection of increasing amounts of TECA dissolved in 0.2 ml of the solvent. A control series of 4 rats received the injection of the same volume of solvent alone. Injections started at day 2 after chamber implantation and were repeated twice a week for 4 weeks. Chambers were collected on day 28 after their insertion, immediately frozen at ­ 80° C and lyophilized. The chamber was, at that time, nearly filled by newly synthetized connective tissue. The dried chamber content was collected, weighed, then dissolved in 0.5M NaOH. An aliquot was taken for DNA measurement according to Fiszer-Safarz et al. [11]. The remaining solution was neutralized with HCl and ethanol was added to the final concentration 80%. After 18 hr at 4° C and a centrifugation at 5,000 g for 30 min, supernatant was collected, the pellet resuspended in normal saline and the ethanol precipitation repeated once. Both supernatants were pooled, evaporated under nitrogen and submitted to acid hydrolysis in 6M HCl at 110° C for 18 hr. This fraction was used for fluorometric measurement of the hydroxyproline contained in the small peptides formed by collagen remodeling in the wound [12]. Ethanol precipitation also eliminated some unidentified pigments that interfered in the protein measurement. The ethanol precipitate was redissolved in 0.5 M NaOH and an aliquot was taken for measurement of total proteins by the method of Lowry [13]. The remaining fraction was collected and used for measurement of collagen and glycosaminoglycans.

Collagen was measured by its hydroxyproline content. An aliquot of the NaOH-solubilized material was neutralized, and hydrolyzed in 6 M HCl at 110° C for 18 hr. Hydroxyproline was quantified in the hydrolyzate by fluorometry [12]. Uronic acid was measured as an index of the glycosaminoglycan content, since it represents 50% of the glycosaminoglycan chains constituting hyaluronic acid, dermatan-sulphate, chondroitin-sulphate and heparan-sulphate. For that purpose, another aliquot was neutralized with acetic acid and digested with pronase in 0.05 M Tris HCl, pH 8.0, 0.02M CaCl₂, for 48 hr at 48° C. Trichloroacetic acid was added to the hydrolyzate to 10% (w/v) final concentration. The samples were centrifuged at 5,000 g for 15 min and the supernatants were dialyzed exhaustively against distilled water at 4° C. Uronic acid was measured in the dialyzate according to Bitter and Muir [14]. The same experiments were done with complete TECA and with its three separated compounds. In this last case, the compounds were used at the same doses as those delivered by TECA injections, knowing that the TECA preparation contained 13.9 % asiatic acid, 35.3% madecassic acid and 38.5% asiaticoside (by weight).

A second set of experiments was performed to study the kinetics of extracellular matrix accumulation in the chambers injected with TECA. For that purpose, wound chambers were inserted on the back of 8 series of 4 rats. A group of 16 rats received the injection of 40 mg TECA dissolved in the usual solvent, twice a week for 4 weeks. The other group of 16 rats received the injection of the solvent alone on the same days. Groups of 8 rats (4 controls and 4 TECA-injected) were then sacrificed on day 7, 14, 21 and 28. Chambers were immediately collected for analysis as described above.

Histological procedures

In some experiments, wound chambers were collected for histological examination. Chambers were immersed immediately after collection in a 10% (by vol.) formalin solution in phosphate buffered saline. The tissues inside were collected with a scalpel. After paraffin inclusion, 4 µm-thick sections were stained with hematoxylin-eosin and safran (saturated solution in absolute ethanol), using a Jung Autosteiner XL apparatus (Leica, Nussloch, Germany). Safran strengthens the staining of collagen fibers [15, 16].

Statistical analysis

Every experiment was done in quadruplicate. Results were expressed as mean ± SEM. Statistical analysis was done by Student's t-test.

Results

Repeated injections of TECA or its separated compounds were totally devoid of any toxicity on the rats. No difference was observed in the growth curves of the different groups of rats (data not shown). Injection of the solvent alone did not accelerate the deposition of extracellular matrix in the wound chambers when compared to non-injected ones (Table I).

Injection of increasing concentrations of TECA twice a week for 4 weeks induced a significant increase of the material deposited in the wound chambers. Dry weight was significantly increased for doses of 10 mg per injection or higher. DNA and total proteins were significantly increased for 20 or 40 mg TECA. Collagen and uronic acid were significantly increased for 40 mg TECA only (Fig. 1).

A kinetic experiment was done in which 40 mg of TECA or the solvent alone (controls) were injected twice a week in the chambers. Groups of 4 rats were sacrificed at days 7, 14, 21 and 28 and the chambers collected for analysis. A significant increase of dry weight, DNA, proteins, collagen and uronic acid was found as soon as day 7 in the TECA-injected rats (Fig. 2). A simple macroscopical observation of the chambers clearly showed that more connective tissue accumulated in the TECA-injected than in the control chambers. Histological examination at day 7 showed that more fibroblasts infiltrated the TECA-injected chambers. At day 28, a large, dense and well organized fibrosis was present in the TECA-injected chambers whereas it was still oedematous and poorly organized in the controls (Fig. 3).

When injected into the chambers, asiatic acid induced a significant increase of the dry weight, DNA, total proteins, collagen, peptidic hydroxyproline and uronic acid contents (Fig. 4). It was more efficient than TECA on total proteins, collagen and uronic acid since significant effects were found at doses as low as 1.5 mg per injection.

Madecassic acid was generally less active than asiatic acid (Fig. 5). Significant increases of dry weight, DNA, total proteins, collagen and uronic acid were found for doses of 7 and 14 mg per injection. In the case of total proteins, however, a biphasic effect was observed with a significant stimulation also observed at the dose 0.875 mg per injection.

The activity of asiaticoside was very different from that of the two other components (Fig. 6). Its activity was generally lower than that of the two previous ones, except for collagen synthesis which was increased by 150% at the dose of 0.96 mg per injection (Fig. 6D). The active doses were also lower than those of asiatic and madecassic acids, with a maximal stimulation at 0.96 mg not only for collagen but also for DNA, total proteins and uronic acid contents. Injection of higher doses did not induce a further increase of any of these parameters.

Discussion

Titrated Extract of Centella Asiatica was reported to stimulate the wound repair process in various types of wounds such as acute surgical wounds [17, 18], bladder ulcers [19], burn wounds [20] and perforating plantar wounds [21]. In this report, we demonstrate that TECA is able to accelerate the formation of new connective tissue in a model of rat experimental wounds. Since TECA is a reconstituted mixture of 3 triterpenes extracted and purified from the plant, we also studied the effects of each of these separated compounds and found that asiatic acid was the most active on the various parameters that we measured.

The model that we used was the wound chamber described by Schilling et al. [10]. In this model, the implanted cylinder is rapidly invaded by inflammatory cells, then fibroblasts, and is nearly filled by newly formed connective tissue after 4 weeks. It enables us to obtain large amounts of granulation tissue at a wound site and to perform static and kinetic studies of the healing process in the absence or presence of pharmacological agents.

Repeated injections of TECA induced an acceleration of healing and a concentration-dependent increase of connective tissue components in the chambers. After 4 weeks, which is near the end of the repair process, dry weight, DNA, total proteins, collagen and uronic acid were all increased in the TECA-injected chambers (40 mg per injection). The stimulation of collagen synthesis was in agreement with previously published data from our laboratory and others, who reported that TECA is able to stimulate collagen synthesis by fibroblast cultures [5-7]. It is also in agreement with previous in vivo studies of Velasco and Romero [22], and Vogel et al. [8] who reported an increased tensile strength of incisional wounds performed in rats treated with TECA. The increase in tensile strength is correlated with an increase of collagen fibers in the wound.

Peptidic hydroxyproline was increased in wound chambers injected with the higher doses of asiatic acid. This demonstrated that collagen turnover and remodeling was accelerated, compared with the control ones. Since collagen content was always higher in the asiatic acid-injected chambers, the activation of collagen synthesis was globally more efficient than catabolism. Wound remodeling is, however, an important feature of the healing process [23].

The stimulation of collagen synthesis in our model was found with complete TECA and with its 3 separated compounds. The active doses were, however, very different since asiatic acid induced a significant stimulation at the dose 1.5 mg (3.0 µmoles) per injection or higher, madecassic acid at 7.0 mg (13.8 µmoles) or higher, and asiaticoside at 0.96 mg (1.0 µmoles). These differences of specific activity, may be linked to structural differences between the 3 triterpenes and/or bioavailability. For instance, asiatic and madecassic acids, two structurally related compounds, differ by a single hydroxy group only [2, 3]. This slight difference seems, however, sufficient to decrease the efficiency of madecassic acid on collagen synthesis by fibroblasts [5]. In the case of asiaticoside, a trisaccharide ester of asiatic acid [3], the presence of the sugar moiety might increase the bioavailability of the compound inside the wound. Some asiaticoside mimetics with a simplified sugar portion were recently shown to possess similar or higher healing properties than the natural compound [9].

Due to their lipidic structure, it is likely that asiatic acid and madecassic acid may penetrate into the cells whereas asiaticoside might be deglycosylated to provide asiatic acid. The intracellular mechanism of action of TECA remains, however, unknown.

The increased DNA content in the TECA-injected chamber was correlated with the higher cell number, especially fibroblasts, clearly visible by histological examination. Since TECA does not stimulate fibroblast proliferation in vitro [5, 7], it is likely that the increased number of cells in the TECA-injected wound chambers depends on a stimulation of cell migration from the surrounding tissues. Another possibility is that TECA might stimulate the expression or activation of some growth factors by the inflammatory cells infiltrated in the wound.

Glycosaminoglycan synthesis, as indicated by uronic acid measurement, was increased in the wound chambers injected by TECA or its purified components. To our knowledge, this is the first report of stimulation of glycosaminoglycan synthesis by TECA. It could be an indirect effect since Tenni et al. reported that TECA was devoid of any effect on glycosaminoglycan synthesis by fibroblast cultures [7]. It is well established that glycosaminoglycans, especially hyaluronic acid and small proteoglycans, play a major role in the healing process and contribute to the organization and strength of the fibrillar network of the wound [24].

Taken together, our data indicate that TECA and its purified components may stimulate the wound healing process. Significant differences of biological activity exist, however, among the 3 triterpenes which are present in the drug. Clinical studies with these purified molecules are necessary to better define their respective therapeutic interest.

CONCLUSION

Acknowledgements

This work was supported by grants from the Université de Reims Champagne-Ardenne, CNRS, and Laboratoires Roche-Nicholas. Mrs Etienne and Deschamps are greatly acknowledged for the preparation of the manuscript, Mrs J. Cornillet-Stoupy for her help in glycosaminoglycan measurements, and Dr. B. Falconnet for providing TECA and its separated compounds.

REFERENCES

1. Clark RAF. The Molecular and Cellular Biology of Wound Repair. New-York, London: Plenum Press, 1996.

2. Polonski J. Sur la constitution chimique de l'acide asiatique, aglycone de l'asiaticoside (3^e mémoire). Rattachement de l'acide asiatique à la série de l'alpha-amyrine; formule développée de l'acide asiatique. Bull Soc Chim France 1953; 173-80.

3. Pinhas H, Billet D, Heitz S, Chaigneau M. Structure de l'acide madécassique, nouveau triterpène de Centella asiatica de Madagascar. Bull Soc Chim France 1967; 1890-5.

4. Polonski J, Sach E, Lederer E. Sur la constitution chimique de la partie glucidique de l'asiaticoside. Bull Soc Chim France 1959; 880-7.

5. Maquart FX, Bellon G, Gillery P, Wegrowski Y, Borel JP. Stimulation of collagen synthesis in fibroblast cultures by a triterpene extracted from Centella asiatica. Connect Tiss Res 1990; 24: 107-20.

6. Boute F, Dumas M, Changague C, Meybeck A. Influence of asiatic acid, madecassic acid, and asiaticoside on human collagen I synthesis. Planta Med 1994; 60: 133-5.

7. Tenni R, Zanaboni G, De Agostini MP, Rossi A, Bendotti C, Cetta G. Effect of the triterpenoid fraction of Centella asiatica on macromolecules of the connective tissue matrix in human skin fibroblast cultures. Ital J Biochem 1988; 37: 69-77.

8. Vogel HG, De Souza NJ, D'Sa A. Effect of terpenoids isolated from Centella asiatica on granuloma tissue. Acta Therap 1990; 16: 285-97.

9. Shim PJ, Park JH, Chang MS, Lim MJ, Kim DH, Jung YH, Jew SS, Park EH, Kim HD. Asiaticoside mimetics as wound healing agents. Bioorg Med Chem Lett 1996; 6: 2937-40.

10. Schilling JA, Joel W, Shurley MT. Wound healing: a comparative study of the histochemical changes in granulation tissue contained in stainless steel wire mesh and polyvinyl sponge cylinders. Surgery 1959; 46: 702-10.

11. Fiszer-Szafarz B, Szafarz D, Guevara de Murillo A. A general, fast and sensitive method for DNA determination: application to rat and mouse liver hepatoma, human leukocytes, chicken fibroblasts and yeast cells. Anal Biochem 1981; 110: 165-70.

12. Bellon G, Malgras A, Randoux A, Borel JP. Further improvement of the fluorometric assay for hydroxyproline. J Chromatogr 1978; 178: 167-72.

13. Lowry OH, Rosebrough NJ, Fair AL, Randall RJ. Protein measurement with the Folin-Phenol reagent. J Biol Chem 1951; 257: 5333-6.

14. Bitter T, Muir HM. A modified uronic acid carbazole reaction. Anal Biochem 1962; 4: 330-4.

15. Snell RS. Clinical and Functional Histology. Boston, Toronto: Little, Brown & Co, 1984.

16. Fanter P, Jolles G. Histochimie Normale et Pathologique. Paris: Gauthier-Villars, 1969.

17. Tsurumi K, Hiramatsu Y, Hayashi M, Fujimura H. Effect of Madecassol (asiaticoside) on wound healing. Pharmacometrics (Tokyo) 1973; 7: 823-43.

18. O'Keefe P. A trial of asiaticoside on skin graft donor areas. Brit J Plast Surg 1974; 27: 194-5.

19. Fam AZ. Use of titrated extract of Centella asiatica (TECA) in bilharzial bladder lesions. Int Surg 1973; 58: 451-5.

20. Monteil R. L'extrait titré de Centella asiatica dans la cicatrisation des brûlures. Feuillets Prat (Paris) 1969; 34:1-16.

21. Nebout M. Resultats d'un essai contrôlé de l'extrait titré de Centella asiatica (ETCA) dans une population lépreuse présentant des maux perforants plantaires. Bull Soc Pathol Exot (Paris) 1974; 5: 471-8.

22. Velasco M, Romero F. Drug interaction between asiaticoside and some antiinflammatory drugs in wound healing of the rat. Curr Therap Res 1976; 19: 121-5.

23. Mignatti P, Rifkin DB, Welgus HG, Parks WC. Proteinases
and tissue remodeling. In: Clark RF, ed. The Molecular and Cellular Biology of Wound Repair. New York, London: Plenum Press, 1996: 427-74.

24. Scott JE. Proteoglycan-fibrillar collagen interactions. Biochem J 1988; 252: 313-23.