Journal of Dermatological Science,
1 (1990) 245-252 245
Elsevier
DESC 00028
Allogeneic cultured epidermal grafts heal chronic ulcers although they do not remain as proved by DNA analysis* Diane I. Roseeuw’, Arlette De Coninck’, Willy Lissens2, Erik Kets4, Inge Liebaers2, A. Vercruysse 3, Yves Vandenberghe Departments
of ‘Dermatology, ‘Genetics and 3Toxicology, Vrije Universiteit Brussel, Brussels, Belgium and 4pepartment Surgery, Military Hospital, Brussels, Belgium
(Received
17 July 1989; accepted
Key words: Cultured keratinocyte
17 January
of
1990)
allografi; Ulcer; DNA analysis
Abstract
To investigate whether allogeneic cultured keratinocytes are rejected or not, and to find out how beneficial their effect on wound healing could be, patients with chronic ulcers were grafted with allogeneic cultured human keratinocytes. In order to examine the epidermal origin of the healed wound, DNA analysis was performed and compared to donor and recipient blood-cell DNA. Healing was observed in 84% of the grafted ulcers by granulation tissue stimulation and wound edge effect. In little time 60”~ of the grafted chronic ulcers healed completely. Although no rejection was observed, DNA analysis revealed that the grafted allogeneic keratinocytes were finally replaced by the patient’s own epidermis. This study confirmed that cultured allogeneic keratinocytes that have been grafted on ulcers, play an important role in the wound healing process.
Introduction Autologous skin grafts have been the treatment of choice for extensive wound closure. However skin replacement, such as in burns, chronic ulcers, Correspondence to: D. Roseeuw, Department of Dermatology, Akademisch Ziekenhuis, Vrije Universiteit Brussel, Laarbeeklaan 101, 1090 Brussels, Belgium. * Part of this work was presented at the ESDR-JSID-SID Tricontinental Meeting in Washington DC., April 26-30, 1989 and published in abstract form in: J Invest Dermatol92: 509, 1989.
0923-181 l/90/$03.50
0 1990 Elsevier Science Publishers
surgical wounds or in skin diseases like epidermolysis bullosa has always been a problem. Indeed, adequate donor sites for autografting are often unavailable, or additional skin injury must be prevented because of the underlying pathological processes. Materials other than the patient’s own skin have been used for temporary wound coverage [ 11. Synthetic skin substitutes as well as cadaver skin allografts from related or unrelated donors seem to be the most effective biologic dressing. However, synthetic skin substitutes have to be
B.V. (Biomedical
Division)
246
regularly replaced and cadaver skin allografts are rejected in about 2 weeks, requiring regrafting or other treatments [ 21. As a new treatment approach, attempts were made to use cultured human epithelium for grafting. The first successfully clinical applications were performed using autologous cultured epithelial grafts for the treatment of burns [ 3-71. Later on, various smaller skin defects were grafted with autologous cultured epidermal cells, with satisfactory results [ 8-101. The major obstacle for the use of cultured autografts, especially in burns, is the 4 weeks waiting time, necessary to grow sufficient quantities of epithelium. A number of authors reported that transplanted cultured allogeneic epidermis was not rejected [ 1 l-131 and might serve as a permanent skin graft. Since allogeneic epithelium can be cultured in advance, skin grafts would be ready, at any time, for grafting. However, data from the literature concerning grafting with allogeneic cultured epidermis are conflicting, although the same culture technique described by Green et al. [ 141 was used. Some authors reported an excellent and permanent take of the allografts [ 1 l-131, while others observed a total and rapid rejection, which actually is postulated to delay the wound healing process [8,15-191. In order to investigate whether allogeneic cultured keratinocytes are, or are not, rejected, and to determine if these allografts are beneficial to wound healing, patients with chronic ulcers were grafted with allogeneic cultured human epithelium. To determine the origin of the epidermis covering the healed ulcers, DNA analysis was performed. The degree of wound healing was evaluated clinically. Materials and Methods Keratinocyte cultures
Human skin was obtained from plastic surgery. All donors were screened for hepatitis, HIV and syphilis. Skin was only taken from healthy patients, under 50 years old. To grow adult human keratinocyte cultures,
the procedure reported by Green [ 141 with minor modifications was used. Strips of keratomed skin (1 x 4 cm) were floated in 0.25 y0 trypsin (Sigma), prepared ex tempore, at 37 “C for 45 min, or left overnight at 4 ’ C to allow dermo-epidermal separation. After inhibition of the trypsin activity by MEM (Gibco) containing 10% foetal calf serum (Gibco) and antibiotics, a cell suspension was obtained by gentle scraping of the epidermis. Feeder cell layer 3T3 fibroblasts were irradiated at 6000 rad and seeded at a density of l-2 x lo4 cells/cm2. The keratinocytes were added at a density of 5 x 104/cm2 in a medium of 75% MEM and 25 y0 M 199 (Gibco). This medium contained 10% foetal calf serum, 0.4 pg/ml hydrocortisone (Roussel), low4 PM cholera toxin (Sigma), 5 pug/ml insulin (Sigma), 7.5 IU/ml Penicillin (Sigma) and 37.5 IU/ml Streptomycin (Sigma). After 2 days the medium was changed and 10 ng/ml epidermal growth factor (Sigma) was added. The cells were grown at 37 “C with 5% CO, in air gassing and the medium was changed every other day. When cultures were optimal for use as grafts, the epithelium was gently lifted from the plate as a whole sheet, using dispase (Boehringer Mannheim) 1.2 units/ml in phosphate buffered saline (PBS). The sheet of cells was backed with a Vaseline gauze or Surfasoft so that the basal layer of the cultured epidermis could easily be placed on the prepared wound bed. Stitches were used to fix the epithelium sheets only in areas of flexibility. Patients and evaluation of healing process
A total of 32 chronic ulcers from 27 patients (male: 12; female: 15) with a mean age of 69 years (range: 38-86 years) were grafted with allogeneic cultured keratinocytes. Chronic ulcers were defined as completely unresponsive to conventional treatment: this means that they remained inert despite an appropriate treatment during the last month before grafting, showing no tendency to progressing granulation tissue formation nor epithelialization. Some ulcers were, under the best possible care, expanding even further. All ulcers were the consequence of a chronic pathologic
241
background such as severe diabetes or arterial and venous insufficiency. The ulcer bed was prepared by gentle debridement and local desinfection until a clean wound was obtained. Occasionally oral antibiotic therapy was given. All ulcers showed, when grafted, a clean non proliferating granulation bed, or reached the level of the fascia. Some were even deeper, down to the underlying tendons or bony structures. When the cultured epithelium sheets were applied to the ulcers, they were covered by a nonadherent dressing (Melolin from Smith and Nephew or Metalline from Lohmann) and supporting bandages. Three to five days after grafting the dressings were removed very carefully. The ulcers were evaluated and dressed again at every 24-48 h interval. Polaroid photographs (l/l) and slides of the ulcers were made every two weeks. The healing activity of these chronic ulcers was evaluated clinically. Since all ulcers included in the study, were refractory to, or even expanding in spite of the conventional treatment, we defined healing activity as: 1. a clear take of the allograft cultures. The presence of tine adherent membranes or epithelial islands in the middle of the ulcer, within the first two weeks after grafting, was considered as clear evidence of a good take. Other conditions for accepting a good take of the graft were the total absence of inflammatory reactions or rejection phenomena such as blister formation. 2. a sudden onset of progressive epithelialization, starting from the wound edges, 5 to 10 days after grafting. 3. an increasing formation of new granulation tissue, tilling up the deep wounds, translating a very intensive angiogenetic activity.
DNA analysis When the ulcers were completely healed a punch biopsy was taken in the middle. The epidermis was separated from the dermis by trypsinization, washed twice in a 0.9% sodium chloride solution and stored at - 80 “C in a minimal volume of the same solution until processing for DNA analysis. DNA analysis was performed to identify the origin of the epidermis. High molecular weight DNA was prepared [ 201 from epidermis and from
venous blood of donor and recipient. Restriction endonuclease digestion of DNA (2 to 5 pg) with Pvull, agarose gel electrophoresis, Southern blotting, hybridisation with the highly polymorphic DNA probe 3’ HVR [21] and autoradiography were carried out as previously described [ 221. Polymorphic patterns, obtained in this way, from epidermis DNA and from DNA isolated from venous blood of donor and recipient were compared. The 3’ HVR probe recognizes a variable number of tandem repeats (VNTR). In practice, this means that, on autoradiography, the intensity of the polymorphic bands will depend on the length of the repeats enclosed by the Pvull restriction sites and that the intensity will decrease with decreasing fragment length. Since the DNA patterns in epidermis could be mixtures of donor and host cells, the sensitivity of our DNA analysis was determined for each recipient and his corresponding donor. This was done in preliminary experiments, by mixing DNA (isolated from blood) of donor and recipient in various proportions and analysed with the 3’HVR probe. These experiments indicated that in the epidermal DNA samples, DNA from both donor and recipient would have been detected at values as low as l-2% of the total DNA. Results A clinical evaluation was done on 27 patients with a total of 32 chronic ulcers. Sixteen ulcers were due to chronic stasis ulceration. Ten were the consequence of severe arterial insufficiency in patients with intermittent claudication and six were due to a pathological background of diabetes mellitus (4 maturity onset diabetes and 2 juvenile diabetes). Of the 32 evaluated ulcers, 3 were to be considered as partial thickness ulcers. Twenty nine ulcers were full thickness wounds of which twenty were, when grafted, covered by a clean granulation tissue bed. Nine of them reached to the level of the muscle fascia or to the deeper lying tendons and bony structures.
248
249 TABLE I Evolution of chronic ulcers after grafting with allogeneic cultured keratinocytes Total ulcers evaluated Number of ulcers %
32 100
Failures
Healing activity
Complete healing
27
19
5
84,3 _
59,3
16
Before grafting all ulcers were free of infection. As shown in Table I, five or 16% of the chronic ulcers did not respond to the cultured allograft treatment. In 27 (84%) of the grafted ulcers a sudden (i.e. 5 to 7 days after grafting) healing activity was expressed. Not only did we observe, like other investigators, an epidermization from the wound edges (Figs. 1,2, 3 and 4) but also an angiogenesis and stimulation of the granulation tissue formation was seen. The healing edges and the skin surrounding the ulcers were very scaly and hyperkeratotic (Figs. 3 and 4). This squamous aspect remained throughout the healing period (Fig. 4) and even up to 2 months thereafter. In nineteen cases, representing 60% of the grafted chronic ulcers, completely healing was observed (Table I). All ulcers with a healing activity (84%) showed an edge effect, expressed by a decreasing diameter (Fig. 4). Only in 12 or 44% of all ulcers could we observe a take of the allografts as postulated by our definition (Table II). From a cosmetic point of view the healed skin looked very satisfactory and strong. When the healing activity of the ulcers after grafting was referred to their background, diabetic ulcers seemed to respond better to the cultured allograft treatment than all others. However a statistic significant difference could not be seen
TABLE II Healing pattern of ulcers after grafting with allogeneic cultured keratinocytes Healing activity
Edge effect
Take of grafts
2 islands Number of ulcers
21
21
12 \ 10 membrane
Y0
100
100
44,4
since not enough ulcers per group were investigated. The healing of the grafted ulcers occured rapidly. The median healing time after grafting was 6 weeks, while the median preexistent duration of the ulcers before grafting was 52 weeks (Table III). Patients with ulcers often complain of severe pain. It was striking how the pain, even in arterial ulcers, disappeared a few hours after grafting. Seventy eight percent of the ulcers were painful at the start of the study. Sixty four percent of these ceased to be painful in the 4 to 24 h after application of the grafts (Table IV).
Fig. 1. Expanding arterial ulcer before grafting with allogeneic cultured epithelium. Fig. 2. Healing activity of cultured epidermal allograft. Five days after grafting is the upper part of the ulcer almost totally healed. Fig. 3. Healing ulcer 2 weeks after grafting. There is a scaly aspect of the surrounding skin. Fig. 4. Edge effect after grafting with cultured allogeneic epidermis. The upper part which healed 3 weeks earlier is very squamous.
250 TABLE
III
TABLE V
Duration time of ulcers before grafting compared to healing time after grafting Number of healed ulcers
Before allograft
After allograft
52 (4-312)
6 (2-17)
a Time is calculated in median values in weeks
TABLE IV Influence of cultured keratinocytes
32 %
of healed ulcers
Number of DNA analysis
Recipient’s DNA
Donor’s DNA
12
12
0
Time”
19
Total
DNA analysis of the epidermis
on pain of ulcers
Chronic ulcers Before graft
After graft
Pain
Pain
No pain
25 (78%) 100
9 36
16 64
Fig. 5. Epidermal DNA analysis of a healed ulcer. The 3’HVR polymorphic patterns of blood DNA’s of the donor and the recipient are on the left and right side, respectively, of the picture. In the middle is the pattern from the epidermis of the healed ulcer of the recipient. This pattern is identical to the recipient’s blood pattern.
The DNA analysis of the epidermis of the healed ulcers was compared to the blood-cell DNA of donor and recipient. An example is shown in Fig. 5. Analysis of epidermis DNA from 12 healed ulcers revealed that this DNA belonged to the recipient and that no donor keratinocytes remained (Table V), demonstrating that there is not a permanent take of the cultured epidermal allografts . Discussion Since the data concerning the take, or rejection, of cultured epidermal allografts were equivocal, we decided to graft chronic ulcers, resistant to conventional treatment, with allogeneic cultured epidermis and to trace the origin of the thicklooking epidermis that covered the healed wounds. Our results indicated that cultured epidermal allografts are very rarely accepted as a whole sheet which can further on develop to a full thickness and normal epidermis. In most cases the grafted sheets were, within 5-14 days after grafting, no longer visible in the wound, although on the edges of these ulcers a hyperkeratotic membrane arised. As shown by DNA analysis of twelve healed ulcers, no donor graft remained. The allogeneic epithelium seemed to ‘vanish’ or dissolve (without rejection symptoms) under the influence of some factor(s) released in the patient’s wound or wound edges. Noteworthy is the sudden response of deep chronic ulcers, that had not healed for years, to the contact with these cultured allogeneic keratinocytes. These cells seemed, before they dis-
251
appeared, to transduce some definite signals or to induce actually unknown changes in 84% of the grafted ulcers. The wound healing capacity, which was apparently definitely lost, could thus be regained. We observed a stimulation of the granulation tissue formation, and like other investigators, a healing from the edges: the so called ‘edge effect’ [ 18,23,24]. This investigation favours the hypothesis that keratinocytes could produce growth factors which stimulate or regulate wound healing. Animal experiments and in vitro-tests support this hypothesis. Barnhill et al. [25] reported a stimulation of the angiogenesis in the chorioallantoic membrane of the chick embryo by the supernatant of conditioned medium wherein keratinocytes had been grown. Eisinger et al. [26] showed that keratinocyte extracts stimulated epidermal cell proliferation in vitro. Probably many other factors play an important role in inducing the healing process by cultured keratinocytes. Wound repair is a very complex process, involving many cell types and interactions between them, involving also different growth factors, biological reactions and mechanical processes like cell-to-cell contact between keratinocytes and between keratinocytes and fibroblasts. In this study a high percent of the ulcers became painless almost immediately after grafting with the cultured keratinocytes. The working mechanism of this effect is unknown. However, it is generally accepted that occluding open wounds from the air with cadaver skin has the best pain relieving effect. There is no reason why these cultured keratinocyte sheets should not be effective in the same way. Speculations can be made that cultured keratinocyte sheets might secrete nerve end influencing factors. We must concede by this study that cultured sheets of allogeneic keratinocytes might be a viable treatment modality of chronic ulcers and a good model for the study of wound healing. Indeed, they provide a good cell-to-cell contact, optimal occlusion of the wound and may release
time and concentration related regulating factors in function of the cell repair necessities. Acknowledgements We would like to thank Dr. Higgs and colleagues for the gift of the 3’HVR probe. This study was supported by a grant from the ‘Nationaal Fonds voor Geneeskundig Wetenschappelijk Onderzoek’ No. 3.0059.87 and from the ‘Onderzoekraad Vrije Universiteit Brussel’. References and uses of 1 Pruitt BA, Levine NS: Characteristics biologic dressings and skin substitutes. Arch Surg 119: 312-322, 1984. 2 Jonker M, Hoogeboom J, van Leeuwen A, Koch CT, van Oud Alblas DB, van Rood JJ: Influence of matching for HLA-DR antigens on skin graft survival. Transplantation 21: 91-94, 1979. 3 O’Connor NE, Mulliken JB, Banks-Schlegel S, Kehinde 0, Green H: Grafting of burns with cultured epithelium prepared from autologous epidermal cells. Lancet i: 75-78, 1981. 4 Teepe RGC, Ponec M, Kreis RW, Hermans RP: Improved grafting method for treatment of burns with autologous cultured human epithelium. Lancet i: 385, 1986. 5 Deglise B. Benathan M, Frenk E, and Krupp S: Resultats preliminaires du traitement des brfiles par autogreffes d’epiderme de culture. Schweiz Med Wochenschr 117: 1380-1383, 1987. 6 Eldad A, Burt A, Clarke JA, Gusterson B: Cultured epithelium as a skin substitute. Burns 13: 173-180, 1987. I Gallico GG, Compton CC, O’Connor NE: New approaches to skin coverage, in Progress in Transplantation. Edited by PJ Morris, NL Tilney, Churchill Livingstone Inc., New York, vol. 2. 1985. 8 Aubdck J, Fritsch P: Autologous versus allogeneic cultivated epidermis for wound grafting, in Proceedings of the International Symposium on clinical use of Cultured Epithelium in Surgery and Dermatology. Edited by RGC Teepe. Medical and Scientific Conferences Ltd., Wheathampstead, Herts, U.K., 1987, pp 59-61. 9 Hefton JM, Caldwell D, Biozes DG, Balin AK, Carter DM: Grafting of skin ulcers with cultured autologous epidermal cells. J Am Acad Dermatol 14: 399-405, 1986. 10 Leigh IM, Purkis PE, Navsaria MA, Phillips TA, Brain A, Hackett MEJ. Cultured epithelium in the treatment of chronic leg ulcers and plastic surgery patients, in Proceedings qf the International Symposium on clinical use
252
11
12
16
17
18
of Cultured Epitheliumin Surgery and Dermatology.Edited by RGC Teepe. Medical and Scientific Conferences Ltd., Wheathampstead, Herts, U.K., 1987, pp 46-52. Hefton JM, Madden MR, Finkelstein JL, Shires GT: Grafting of burn patients with allografts of cultured epidermal cells. Lancet ii: 428-430, 1983. Thivolet J, Faure M, Demidem A, Mauduit G: Long term survival and immunological tolerance of human epiderma1 allografts produced in culture. Transplantation 42: 274-280, 1986. Thivolet J, Faure M, Demidem A, Mauduit G: Cultured human epidermal allografts are not rejected for a long period. Arch Dermatol Res 278: 252-254, 1986. Green H, Kehinde 0, Thomas J: Growth of cultured human epidermal cells into multiple epithelia suitable for grafting. Proc Nat1 Acad Sci USA 76: 5665-5668, 1979. Clarke JA, Burt AM, Eldad A, Yardeni P, Gusterson B: Allograft cultured epithelium and meshed allograft skin, in Proceedings of the InternationalSymposium on the use of CulturedEpitheliumin Surgery and Dermatology.Edited by RGC Teepe. Medical and Scientific Conferences Ltd., Wheathampstead, Herts, U.K., 1987, pp 31-34. Roseeuw D, De Coninck A, Vandenberghe Y, Lissens W, Kets E: Traitement dun ulctre par allogreffe d’tpiderme humain cultive. Nouvelles Dermatologiques 6: 504,1987. Roseeuw D, Lissens W, De Coninck A, Vandenberghe Y, Liebaers I: Epidermal DNA analysis of chronic ulcers grafted with allogenic cultured epidermis. J Invest Dermatol 92: 509, 1989.A. Aubock J, Irschick E, Romani N, Kompatscher P, Hopfl. R, Herold M, Schuler G, Bauer M, Huber C, Fritsch P: Rejection after a slightly prolonged survival
time of Langerhans cell-free allogenic cultured epidermis used for wound coverage in humans. Transplantation 45: 730-737, 1988. 19 Gielen V, Faure M, Mauduit G, Thivolet J: Progressive replacement of human cultured epithelial allografts by recipient cells as evidenced by HLA-class I antigen expression. Dermatologica 175: 166-170, 1987. 20 Kunkel L, Smith K, Boyer S, Borgaonkar D, Wachtel S, Miller 0, Berg W, Jones H, Rary J: Analysis of human Y-chromosome specific reiterated DNA in chromosome variants. Proc Nat1 Acad Sci USA 74: 1245-1249, 1977. 21 Jarman A, Nicholls R, Weatherall J, Clegg J, Higgs D: Molecular characterisation of a hypervariable region downstream of the human alfa-globin gene cluster. EMBO J5 1857-1863, 1986. 22 Maniatis T, Fritsch E, Sambrook J: Molecular cloning A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1982. 23 Leigh IM, Purkis PE, Navsaria HA, Phillips TJ: Treatment of chronic venous ulcers with sheets of cultured allogenic keratinocytes. Br J Dermatol 117, 591-597, 1987. 24 Clancy JMP, Shehade SA, Blight AE, Young KE, Levick PL: Treatment of leg ulcers with cultured epithelial allografts. J Am Acad Dermatol 18: 1356-1357, 1988. 25 Barnhill RL, Parkinson EK, Ryan TJ: Supernatants from cultured human epidermal keratinocytes stimulate angiogenesis. Br J Dermatol 110: 273-281, 1984. 26 Eisinger M, Sadan S, Silver IA, Flick RB: Growth regulation of skin cells by epidermal cell-derived factors: implications for wound healing. Proc Nat1 Acad Sci USA 85: 1937-1941, 1988.
1 (1990) 245-252 245
Elsevier
DESC 00028
Allogeneic cultured epidermal grafts heal chronic ulcers although they do not remain as proved by DNA analysis* Diane I. Roseeuw’, Arlette De Coninck’, Willy Lissens2, Erik Kets4, Inge Liebaers2, A. Vercruysse 3, Yves Vandenberghe Departments
of ‘Dermatology, ‘Genetics and 3Toxicology, Vrije Universiteit Brussel, Brussels, Belgium and 4pepartment Surgery, Military Hospital, Brussels, Belgium
(Received
17 July 1989; accepted
Key words: Cultured keratinocyte
17 January
of
1990)
allografi; Ulcer; DNA analysis
Abstract
To investigate whether allogeneic cultured keratinocytes are rejected or not, and to find out how beneficial their effect on wound healing could be, patients with chronic ulcers were grafted with allogeneic cultured human keratinocytes. In order to examine the epidermal origin of the healed wound, DNA analysis was performed and compared to donor and recipient blood-cell DNA. Healing was observed in 84% of the grafted ulcers by granulation tissue stimulation and wound edge effect. In little time 60”~ of the grafted chronic ulcers healed completely. Although no rejection was observed, DNA analysis revealed that the grafted allogeneic keratinocytes were finally replaced by the patient’s own epidermis. This study confirmed that cultured allogeneic keratinocytes that have been grafted on ulcers, play an important role in the wound healing process.
Introduction Autologous skin grafts have been the treatment of choice for extensive wound closure. However skin replacement, such as in burns, chronic ulcers, Correspondence to: D. Roseeuw, Department of Dermatology, Akademisch Ziekenhuis, Vrije Universiteit Brussel, Laarbeeklaan 101, 1090 Brussels, Belgium. * Part of this work was presented at the ESDR-JSID-SID Tricontinental Meeting in Washington DC., April 26-30, 1989 and published in abstract form in: J Invest Dermatol92: 509, 1989.
0923-181 l/90/$03.50
0 1990 Elsevier Science Publishers
surgical wounds or in skin diseases like epidermolysis bullosa has always been a problem. Indeed, adequate donor sites for autografting are often unavailable, or additional skin injury must be prevented because of the underlying pathological processes. Materials other than the patient’s own skin have been used for temporary wound coverage [ 11. Synthetic skin substitutes as well as cadaver skin allografts from related or unrelated donors seem to be the most effective biologic dressing. However, synthetic skin substitutes have to be
B.V. (Biomedical
Division)
246
regularly replaced and cadaver skin allografts are rejected in about 2 weeks, requiring regrafting or other treatments [ 21. As a new treatment approach, attempts were made to use cultured human epithelium for grafting. The first successfully clinical applications were performed using autologous cultured epithelial grafts for the treatment of burns [ 3-71. Later on, various smaller skin defects were grafted with autologous cultured epidermal cells, with satisfactory results [ 8-101. The major obstacle for the use of cultured autografts, especially in burns, is the 4 weeks waiting time, necessary to grow sufficient quantities of epithelium. A number of authors reported that transplanted cultured allogeneic epidermis was not rejected [ 1 l-131 and might serve as a permanent skin graft. Since allogeneic epithelium can be cultured in advance, skin grafts would be ready, at any time, for grafting. However, data from the literature concerning grafting with allogeneic cultured epidermis are conflicting, although the same culture technique described by Green et al. [ 141 was used. Some authors reported an excellent and permanent take of the allografts [ 1 l-131, while others observed a total and rapid rejection, which actually is postulated to delay the wound healing process [8,15-191. In order to investigate whether allogeneic cultured keratinocytes are, or are not, rejected, and to determine if these allografts are beneficial to wound healing, patients with chronic ulcers were grafted with allogeneic cultured human epithelium. To determine the origin of the epidermis covering the healed ulcers, DNA analysis was performed. The degree of wound healing was evaluated clinically. Materials and Methods Keratinocyte cultures
Human skin was obtained from plastic surgery. All donors were screened for hepatitis, HIV and syphilis. Skin was only taken from healthy patients, under 50 years old. To grow adult human keratinocyte cultures,
the procedure reported by Green [ 141 with minor modifications was used. Strips of keratomed skin (1 x 4 cm) were floated in 0.25 y0 trypsin (Sigma), prepared ex tempore, at 37 “C for 45 min, or left overnight at 4 ’ C to allow dermo-epidermal separation. After inhibition of the trypsin activity by MEM (Gibco) containing 10% foetal calf serum (Gibco) and antibiotics, a cell suspension was obtained by gentle scraping of the epidermis. Feeder cell layer 3T3 fibroblasts were irradiated at 6000 rad and seeded at a density of l-2 x lo4 cells/cm2. The keratinocytes were added at a density of 5 x 104/cm2 in a medium of 75% MEM and 25 y0 M 199 (Gibco). This medium contained 10% foetal calf serum, 0.4 pg/ml hydrocortisone (Roussel), low4 PM cholera toxin (Sigma), 5 pug/ml insulin (Sigma), 7.5 IU/ml Penicillin (Sigma) and 37.5 IU/ml Streptomycin (Sigma). After 2 days the medium was changed and 10 ng/ml epidermal growth factor (Sigma) was added. The cells were grown at 37 “C with 5% CO, in air gassing and the medium was changed every other day. When cultures were optimal for use as grafts, the epithelium was gently lifted from the plate as a whole sheet, using dispase (Boehringer Mannheim) 1.2 units/ml in phosphate buffered saline (PBS). The sheet of cells was backed with a Vaseline gauze or Surfasoft so that the basal layer of the cultured epidermis could easily be placed on the prepared wound bed. Stitches were used to fix the epithelium sheets only in areas of flexibility. Patients and evaluation of healing process
A total of 32 chronic ulcers from 27 patients (male: 12; female: 15) with a mean age of 69 years (range: 38-86 years) were grafted with allogeneic cultured keratinocytes. Chronic ulcers were defined as completely unresponsive to conventional treatment: this means that they remained inert despite an appropriate treatment during the last month before grafting, showing no tendency to progressing granulation tissue formation nor epithelialization. Some ulcers were, under the best possible care, expanding even further. All ulcers were the consequence of a chronic pathologic
241
background such as severe diabetes or arterial and venous insufficiency. The ulcer bed was prepared by gentle debridement and local desinfection until a clean wound was obtained. Occasionally oral antibiotic therapy was given. All ulcers showed, when grafted, a clean non proliferating granulation bed, or reached the level of the fascia. Some were even deeper, down to the underlying tendons or bony structures. When the cultured epithelium sheets were applied to the ulcers, they were covered by a nonadherent dressing (Melolin from Smith and Nephew or Metalline from Lohmann) and supporting bandages. Three to five days after grafting the dressings were removed very carefully. The ulcers were evaluated and dressed again at every 24-48 h interval. Polaroid photographs (l/l) and slides of the ulcers were made every two weeks. The healing activity of these chronic ulcers was evaluated clinically. Since all ulcers included in the study, were refractory to, or even expanding in spite of the conventional treatment, we defined healing activity as: 1. a clear take of the allograft cultures. The presence of tine adherent membranes or epithelial islands in the middle of the ulcer, within the first two weeks after grafting, was considered as clear evidence of a good take. Other conditions for accepting a good take of the graft were the total absence of inflammatory reactions or rejection phenomena such as blister formation. 2. a sudden onset of progressive epithelialization, starting from the wound edges, 5 to 10 days after grafting. 3. an increasing formation of new granulation tissue, tilling up the deep wounds, translating a very intensive angiogenetic activity.
DNA analysis When the ulcers were completely healed a punch biopsy was taken in the middle. The epidermis was separated from the dermis by trypsinization, washed twice in a 0.9% sodium chloride solution and stored at - 80 “C in a minimal volume of the same solution until processing for DNA analysis. DNA analysis was performed to identify the origin of the epidermis. High molecular weight DNA was prepared [ 201 from epidermis and from
venous blood of donor and recipient. Restriction endonuclease digestion of DNA (2 to 5 pg) with Pvull, agarose gel electrophoresis, Southern blotting, hybridisation with the highly polymorphic DNA probe 3’ HVR [21] and autoradiography were carried out as previously described [ 221. Polymorphic patterns, obtained in this way, from epidermis DNA and from DNA isolated from venous blood of donor and recipient were compared. The 3’ HVR probe recognizes a variable number of tandem repeats (VNTR). In practice, this means that, on autoradiography, the intensity of the polymorphic bands will depend on the length of the repeats enclosed by the Pvull restriction sites and that the intensity will decrease with decreasing fragment length. Since the DNA patterns in epidermis could be mixtures of donor and host cells, the sensitivity of our DNA analysis was determined for each recipient and his corresponding donor. This was done in preliminary experiments, by mixing DNA (isolated from blood) of donor and recipient in various proportions and analysed with the 3’HVR probe. These experiments indicated that in the epidermal DNA samples, DNA from both donor and recipient would have been detected at values as low as l-2% of the total DNA. Results A clinical evaluation was done on 27 patients with a total of 32 chronic ulcers. Sixteen ulcers were due to chronic stasis ulceration. Ten were the consequence of severe arterial insufficiency in patients with intermittent claudication and six were due to a pathological background of diabetes mellitus (4 maturity onset diabetes and 2 juvenile diabetes). Of the 32 evaluated ulcers, 3 were to be considered as partial thickness ulcers. Twenty nine ulcers were full thickness wounds of which twenty were, when grafted, covered by a clean granulation tissue bed. Nine of them reached to the level of the muscle fascia or to the deeper lying tendons and bony structures.
248
249 TABLE I Evolution of chronic ulcers after grafting with allogeneic cultured keratinocytes Total ulcers evaluated Number of ulcers %
32 100
Failures
Healing activity
Complete healing
27
19
5
84,3 _
59,3
16
Before grafting all ulcers were free of infection. As shown in Table I, five or 16% of the chronic ulcers did not respond to the cultured allograft treatment. In 27 (84%) of the grafted ulcers a sudden (i.e. 5 to 7 days after grafting) healing activity was expressed. Not only did we observe, like other investigators, an epidermization from the wound edges (Figs. 1,2, 3 and 4) but also an angiogenesis and stimulation of the granulation tissue formation was seen. The healing edges and the skin surrounding the ulcers were very scaly and hyperkeratotic (Figs. 3 and 4). This squamous aspect remained throughout the healing period (Fig. 4) and even up to 2 months thereafter. In nineteen cases, representing 60% of the grafted chronic ulcers, completely healing was observed (Table I). All ulcers with a healing activity (84%) showed an edge effect, expressed by a decreasing diameter (Fig. 4). Only in 12 or 44% of all ulcers could we observe a take of the allografts as postulated by our definition (Table II). From a cosmetic point of view the healed skin looked very satisfactory and strong. When the healing activity of the ulcers after grafting was referred to their background, diabetic ulcers seemed to respond better to the cultured allograft treatment than all others. However a statistic significant difference could not be seen
TABLE II Healing pattern of ulcers after grafting with allogeneic cultured keratinocytes Healing activity
Edge effect
Take of grafts
2 islands Number of ulcers
21
21
12 \ 10 membrane
Y0
100
100
44,4
since not enough ulcers per group were investigated. The healing of the grafted ulcers occured rapidly. The median healing time after grafting was 6 weeks, while the median preexistent duration of the ulcers before grafting was 52 weeks (Table III). Patients with ulcers often complain of severe pain. It was striking how the pain, even in arterial ulcers, disappeared a few hours after grafting. Seventy eight percent of the ulcers were painful at the start of the study. Sixty four percent of these ceased to be painful in the 4 to 24 h after application of the grafts (Table IV).
Fig. 1. Expanding arterial ulcer before grafting with allogeneic cultured epithelium. Fig. 2. Healing activity of cultured epidermal allograft. Five days after grafting is the upper part of the ulcer almost totally healed. Fig. 3. Healing ulcer 2 weeks after grafting. There is a scaly aspect of the surrounding skin. Fig. 4. Edge effect after grafting with cultured allogeneic epidermis. The upper part which healed 3 weeks earlier is very squamous.
250 TABLE
III
TABLE V
Duration time of ulcers before grafting compared to healing time after grafting Number of healed ulcers
Before allograft
After allograft
52 (4-312)
6 (2-17)
a Time is calculated in median values in weeks
TABLE IV Influence of cultured keratinocytes
32 %
of healed ulcers
Number of DNA analysis
Recipient’s DNA
Donor’s DNA
12
12
0
Time”
19
Total
DNA analysis of the epidermis
on pain of ulcers
Chronic ulcers Before graft
After graft
Pain
Pain
No pain
25 (78%) 100
9 36
16 64
Fig. 5. Epidermal DNA analysis of a healed ulcer. The 3’HVR polymorphic patterns of blood DNA’s of the donor and the recipient are on the left and right side, respectively, of the picture. In the middle is the pattern from the epidermis of the healed ulcer of the recipient. This pattern is identical to the recipient’s blood pattern.
The DNA analysis of the epidermis of the healed ulcers was compared to the blood-cell DNA of donor and recipient. An example is shown in Fig. 5. Analysis of epidermis DNA from 12 healed ulcers revealed that this DNA belonged to the recipient and that no donor keratinocytes remained (Table V), demonstrating that there is not a permanent take of the cultured epidermal allografts . Discussion Since the data concerning the take, or rejection, of cultured epidermal allografts were equivocal, we decided to graft chronic ulcers, resistant to conventional treatment, with allogeneic cultured epidermis and to trace the origin of the thicklooking epidermis that covered the healed wounds. Our results indicated that cultured epidermal allografts are very rarely accepted as a whole sheet which can further on develop to a full thickness and normal epidermis. In most cases the grafted sheets were, within 5-14 days after grafting, no longer visible in the wound, although on the edges of these ulcers a hyperkeratotic membrane arised. As shown by DNA analysis of twelve healed ulcers, no donor graft remained. The allogeneic epithelium seemed to ‘vanish’ or dissolve (without rejection symptoms) under the influence of some factor(s) released in the patient’s wound or wound edges. Noteworthy is the sudden response of deep chronic ulcers, that had not healed for years, to the contact with these cultured allogeneic keratinocytes. These cells seemed, before they dis-
251
appeared, to transduce some definite signals or to induce actually unknown changes in 84% of the grafted ulcers. The wound healing capacity, which was apparently definitely lost, could thus be regained. We observed a stimulation of the granulation tissue formation, and like other investigators, a healing from the edges: the so called ‘edge effect’ [ 18,23,24]. This investigation favours the hypothesis that keratinocytes could produce growth factors which stimulate or regulate wound healing. Animal experiments and in vitro-tests support this hypothesis. Barnhill et al. [25] reported a stimulation of the angiogenesis in the chorioallantoic membrane of the chick embryo by the supernatant of conditioned medium wherein keratinocytes had been grown. Eisinger et al. [26] showed that keratinocyte extracts stimulated epidermal cell proliferation in vitro. Probably many other factors play an important role in inducing the healing process by cultured keratinocytes. Wound repair is a very complex process, involving many cell types and interactions between them, involving also different growth factors, biological reactions and mechanical processes like cell-to-cell contact between keratinocytes and between keratinocytes and fibroblasts. In this study a high percent of the ulcers became painless almost immediately after grafting with the cultured keratinocytes. The working mechanism of this effect is unknown. However, it is generally accepted that occluding open wounds from the air with cadaver skin has the best pain relieving effect. There is no reason why these cultured keratinocyte sheets should not be effective in the same way. Speculations can be made that cultured keratinocyte sheets might secrete nerve end influencing factors. We must concede by this study that cultured sheets of allogeneic keratinocytes might be a viable treatment modality of chronic ulcers and a good model for the study of wound healing. Indeed, they provide a good cell-to-cell contact, optimal occlusion of the wound and may release
time and concentration related regulating factors in function of the cell repair necessities. Acknowledgements We would like to thank Dr. Higgs and colleagues for the gift of the 3’HVR probe. This study was supported by a grant from the ‘Nationaal Fonds voor Geneeskundig Wetenschappelijk Onderzoek’ No. 3.0059.87 and from the ‘Onderzoekraad Vrije Universiteit Brussel’. References and uses of 1 Pruitt BA, Levine NS: Characteristics biologic dressings and skin substitutes. Arch Surg 119: 312-322, 1984. 2 Jonker M, Hoogeboom J, van Leeuwen A, Koch CT, van Oud Alblas DB, van Rood JJ: Influence of matching for HLA-DR antigens on skin graft survival. Transplantation 21: 91-94, 1979. 3 O’Connor NE, Mulliken JB, Banks-Schlegel S, Kehinde 0, Green H: Grafting of burns with cultured epithelium prepared from autologous epidermal cells. Lancet i: 75-78, 1981. 4 Teepe RGC, Ponec M, Kreis RW, Hermans RP: Improved grafting method for treatment of burns with autologous cultured human epithelium. Lancet i: 385, 1986. 5 Deglise B. Benathan M, Frenk E, and Krupp S: Resultats preliminaires du traitement des brfiles par autogreffes d’epiderme de culture. Schweiz Med Wochenschr 117: 1380-1383, 1987. 6 Eldad A, Burt A, Clarke JA, Gusterson B: Cultured epithelium as a skin substitute. Burns 13: 173-180, 1987. I Gallico GG, Compton CC, O’Connor NE: New approaches to skin coverage, in Progress in Transplantation. Edited by PJ Morris, NL Tilney, Churchill Livingstone Inc., New York, vol. 2. 1985. 8 Aubdck J, Fritsch P: Autologous versus allogeneic cultivated epidermis for wound grafting, in Proceedings of the International Symposium on clinical use of Cultured Epithelium in Surgery and Dermatology. Edited by RGC Teepe. Medical and Scientific Conferences Ltd., Wheathampstead, Herts, U.K., 1987, pp 59-61. 9 Hefton JM, Caldwell D, Biozes DG, Balin AK, Carter DM: Grafting of skin ulcers with cultured autologous epidermal cells. J Am Acad Dermatol 14: 399-405, 1986. 10 Leigh IM, Purkis PE, Navsaria MA, Phillips TA, Brain A, Hackett MEJ. Cultured epithelium in the treatment of chronic leg ulcers and plastic surgery patients, in Proceedings qf the International Symposium on clinical use
252
11
12
16
17
18
of Cultured Epitheliumin Surgery and Dermatology.Edited by RGC Teepe. Medical and Scientific Conferences Ltd., Wheathampstead, Herts, U.K., 1987, pp 46-52. Hefton JM, Madden MR, Finkelstein JL, Shires GT: Grafting of burn patients with allografts of cultured epidermal cells. Lancet ii: 428-430, 1983. Thivolet J, Faure M, Demidem A, Mauduit G: Long term survival and immunological tolerance of human epiderma1 allografts produced in culture. Transplantation 42: 274-280, 1986. Thivolet J, Faure M, Demidem A, Mauduit G: Cultured human epidermal allografts are not rejected for a long period. Arch Dermatol Res 278: 252-254, 1986. Green H, Kehinde 0, Thomas J: Growth of cultured human epidermal cells into multiple epithelia suitable for grafting. Proc Nat1 Acad Sci USA 76: 5665-5668, 1979. Clarke JA, Burt AM, Eldad A, Yardeni P, Gusterson B: Allograft cultured epithelium and meshed allograft skin, in Proceedings of the InternationalSymposium on the use of CulturedEpitheliumin Surgery and Dermatology.Edited by RGC Teepe. Medical and Scientific Conferences Ltd., Wheathampstead, Herts, U.K., 1987, pp 31-34. Roseeuw D, De Coninck A, Vandenberghe Y, Lissens W, Kets E: Traitement dun ulctre par allogreffe d’tpiderme humain cultive. Nouvelles Dermatologiques 6: 504,1987. Roseeuw D, Lissens W, De Coninck A, Vandenberghe Y, Liebaers I: Epidermal DNA analysis of chronic ulcers grafted with allogenic cultured epidermis. J Invest Dermatol 92: 509, 1989.A. Aubock J, Irschick E, Romani N, Kompatscher P, Hopfl. R, Herold M, Schuler G, Bauer M, Huber C, Fritsch P: Rejection after a slightly prolonged survival
time of Langerhans cell-free allogenic cultured epidermis used for wound coverage in humans. Transplantation 45: 730-737, 1988. 19 Gielen V, Faure M, Mauduit G, Thivolet J: Progressive replacement of human cultured epithelial allografts by recipient cells as evidenced by HLA-class I antigen expression. Dermatologica 175: 166-170, 1987. 20 Kunkel L, Smith K, Boyer S, Borgaonkar D, Wachtel S, Miller 0, Berg W, Jones H, Rary J: Analysis of human Y-chromosome specific reiterated DNA in chromosome variants. Proc Nat1 Acad Sci USA 74: 1245-1249, 1977. 21 Jarman A, Nicholls R, Weatherall J, Clegg J, Higgs D: Molecular characterisation of a hypervariable region downstream of the human alfa-globin gene cluster. EMBO J5 1857-1863, 1986. 22 Maniatis T, Fritsch E, Sambrook J: Molecular cloning A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1982. 23 Leigh IM, Purkis PE, Navsaria HA, Phillips TJ: Treatment of chronic venous ulcers with sheets of cultured allogenic keratinocytes. Br J Dermatol 117, 591-597, 1987. 24 Clancy JMP, Shehade SA, Blight AE, Young KE, Levick PL: Treatment of leg ulcers with cultured epithelial allografts. J Am Acad Dermatol 18: 1356-1357, 1988. 25 Barnhill RL, Parkinson EK, Ryan TJ: Supernatants from cultured human epidermal keratinocytes stimulate angiogenesis. Br J Dermatol 110: 273-281, 1984. 26 Eisinger M, Sadan S, Silver IA, Flick RB: Growth regulation of skin cells by epidermal cell-derived factors: implications for wound healing. Proc Nat1 Acad Sci USA 85: 1937-1941, 1988.
